Entanglements of Modernity, Colonialism and Genocide Burundi and Rwanda in Historical-Sociological Perspective

Home , Abiru, Cyprien Ntaryamira, Genocide justification, Genocide prevention, Geography of Burundi, History of Burundi

UNIVERSITY OF LEEDS

Jack Dominic Palmer University of Leeds School of Sociology and Social Policy

January 2017

Submitted in accordance with the requirements for the degree of Doctor of Philosophy ii

The candidate confirms that the work submitted is their own and that appropriate credit has been given where reference has been made to the work of others. This copy has been supplied on the understanding that it is copyright material and that no quotation from the thesis may be published without proper acknowledgement.

The right of Jack Dominic Palmer to be identified as Author of this work has been asserted by Jack Dominic Palmer in accordance with the Copyright, Designs and Patents Act 1988.

iii

ACKNOWLEDGEMENTS

I would firstly like to thank Dr Mark Davis and Dr Tom Campbell. The quality of their guidance, insight and friendship has been a huge source of support and has helped me through tough periods in which my motivation and enthusiasm for the project were tested to their limits. I drew great inspiration from the insightful and constructive critical comments and recommendations of Dr Shirley Tate and Dr Austin Harrington when the thesis was at the upgrade stage, and I am also grateful for generous follow-up discussions with the latter. I am very appreciative of the staff members in SSP with whom I have worked closely in my teaching capacities, as well as of the staff in the office who do such a great job at holding the department together. I would also like to say a huge thanks to my friends Emma Bimpson, Ben Hirst, Robert Lee, Emily Maddox and Jenny Read for their companionship, humour and support. There are things I won’t miss about doing a PhD but I will certainly miss seeing you all regularly. As ever, I am grateful for the unconditional love and support of my family in the South: Mum, Dad, Luke, Holly, Tegan, Tasmin and the newest addition, Aurelia. I am grateful too to my surrogate West Yorkshire family, the Stokoes. I am also extremely thankful to Muhammad Sajjad, whose kindness and dignity in spite of everything he’s been through have been sobering and inspiring. I am very grateful that mine and Jean-Marie Ninziza’s biographies became entangled here in Leeds, and that he traced his so candidly on an unforgettable trip around Burundi in 2013. I thank him and Joy for their hospitality, and all who greeted and hosted us so warmly. I also thank JM for his intelligent and engaging conversation, which has had a profound bearing on the direction of this project - agashingura cumu! Finally, I am more appreciative than words can express for the love, support and intellect of my partner, Yoshiko. She has grounded me in my flighty moments and pulled me out of pits of despair. This thesis is dedicated to her.

ABSTRACT

This thesis explores two lines of critique of social theories of modernity by way of a historical sociological analysis of Burundi and Rwanda. Firstly, it engages with arguments about the Euro-/Western-centric assumptions which are suggested to have underpinned many conceptualisations of modernity. Secondly, it considers the notion that the processes of modernity move gradually, if precariously, towards more peaceable forms of cohabitation within and between societies. In doing so, it draws on and develops the theoretical framework of entanglement, which emphasises the existence of a variety of intertwined historical routes to and through modernity. Central to the analysis is a critique of both the idea that modernity entails a progressive ‘detraditionalisation’ or destruction of traditional societal forms, and the idea that tradition provides a repository of cultural resources upon which are founded distinct, plural ‘modernities’. In the case of Burundi and Rwanda, I argue that colonial modernity, in its indirect rule format, in important respects ‘solidified’ tradition in racial terms. In the transition to independence, the colonial legacy both enabled and delimited autonomous societal self- understandings and political movements. In the postcolonial period, the tension between the modern commitment to autonomy on the one hand and seemingly traditional legacies on the other has been realised in profoundly destructive and violent ways. I conclude that the historical experiences of extremely violent social conflict in Burundi and Rwanda are situated within a specific route to and through modernity. The original contribution of the thesis is twofold. Firstly, it presents a new substantive case study to the analysis of non-Western experiences and interpretations of modernity. Secondly, in doing so, it offers a theoretical contribution to debates concerning the multiplicity of modernity which have arisen principally in the paradigm of ‘multiple modernities’ and postcolonial sociology.

Contents

Acknowledgements ...... iii Abstract ...... iv Glossary of terms and acronyms ...... vii PART ONE ...... 1 CHAPTER ONE – INTRODUCTION ...... 1 Original Contribution ...... 2 Audience ...... 4 Why Burundi and Rwanda? ...... 5 Research Questions ...... 9 Structure of the Thesis ...... 9 Chapter Synopsis ...... 11 CHAPTER TWO – GENOCIDE AND COLONIALISM: PRELIMINARY CONCEPTUAL AND DEFINITIONAL PROBLEMS...... 15 Genocide ...... 15 Colonialism ...... 34 Genocide and Colonialism ...... 42 CHAPTER THREE – THEORISING THE MULTIPLICITY OF MODERNITY: TOWARDS ENTANGLED HISTORICAL ROUTES TO AND THROUGH MODERNITY ...... 48 Modernity, Modernisation, Modern Society ...... 49 Modernity, Colonialism and Eurocentrism ...... 52 Modernity as a Time-Orientation ...... 55 Multiple Modernities ...... 59 Entangled Historical Routes to and through Modernity ...... 68 CHAPTER FOUR – UNDERSTANDING ENTANGLEMENTS: HERMENEUTICS, HISTORICAL SOCIOLOGY, AND METHOD ...... 82 Problems of Global and Historical Understanding ...... 83 Philosophical Hermeneutics, Historical Sociology and Area Studies ...... 86 Using Primary and Secondary Documentary Sources ...... 98 Limitations, Reflexivity and Ethics ...... 105 PART TWO ...... 108 CHAPTER FIVE – PRECOLONIAL BURUNDI AND RWANDA: A HISTORICAL SURVEY ...... 108 The Great Lakes Region ...... 110 Rwanda ...... 115 Burundi ...... 124 Intermediate Reflections ...... 129 vi

CHAPTER SIX – THE COLONIAL ENTANGLEMENT, 1905-1945: THE RACIALISATION OF TRADITION...... 132 Theorising the Colonial State ...... 134 The Hamitic Hypothesis and European Exploration ...... 138 Situating Colonial Rule ...... 142 The Colonial Reconstruction ...... 153 Intermediate Reflections ...... 161 CHAPTER SEVEN – TRAJECTORIES TOWARDS INDEPENDENCE, 1945-1965: MULTIPLE ‘SOCIETAL SELF-UNDERSTANDINGS’ ...... 166 Multiplicity, Elites, and Societal Self-Understandings ...... 168 Situating Burundi and Rwanda in Transformations of the Geo-Historical Entanglements of Modernity ...... 176 Disentangling Trajectories towards Independence ...... 181 Rwanda ...... 181 Burundi ...... 191 Intermediate Reflections ...... 199 CHAPTER EIGHT – POSTCOLONIAL CRISIS AND GENOCIDE, 1965-1994: TRAUMAS OF MODERNITY...... 205 The Modernity of Genocide ...... 208 Disentangling the ‘Manifold Interlockings’ of Post-Independence Violence ...... 213 Rwanda ...... 214 Burundi ...... 227 Intermediate Reflections ...... 238 PART THREE ...... 246 CHAPTER NINE – CONCLUSION ...... 246 Original Contributions ...... 246 Future Research Agenda ...... 257 Implications for Contemporary Developments ...... 259 Bibliography ...... 268

vii

GLOSSARY OF TERMS AND ACRONYMS

Abashingatahe – Judicial interpreters in precolonial Burundi. Abiru – Narrators of history at the Nyiginya court in Rwanda. Banyamabanga – Custodians of court rituals in the Burundi kingdom. Bitekerezo – Historical oral literature and poetry promulgated at the Nyiginya court in Rwanda. Colline – Hill, the basic spatial unit of social life in Burundi and Rwanda. Front pour la Démocratie au Burundi (FRODEBU) – Burundian political party led by Melchior Ndadaye elected in 1993. Ganwa – ‘Princes of the blood’; political rulers in precolonial and colonial Burundi divided into distinct lineages descending from the mwami (Batare, Bezi, Bataga and Bambutsa). Ikinimicu – Theatrical practice in Rwanda. Imana – Often translated as God (shared by both Burundians and Rwandans) though more accurately represented as a transformative cosmic force. impuzamugambi – Paramilitary death squad during the 1994 genocide in Rwanda, translating as ‘those with the same goal’. Interahamwe – Paramilitary death squad during the 1994 genocide in Rwanda, translating as ‘those who stand together’. Jeunesse Nationaliste Rwagasore (JNR, later JRR) – Youth wing of UPRONA in Burundi. Kangura – Rwandan magazine which became infamous for its propaganda leading up to 1994 genocide, title translating as ‘wake others up’. Kwihutura – Process in precolonial Rwanda whereby Hutu could become Tutsi (the opposite processes was named gucupira). Mouvement Révolutionnaire pour le Développement (MRND) – Single political party of the ‘second republic’ in Rwanda, led by Juvenal Habyarimana. Mwami – King in Kirundi and Kinyarwanda. Nyabingi cult – Millenarian protest movement which emerged in northwestern Rwanda and southwestern Uganda in response to precolonial Nyiginya expansion and the centralisation processes that accompanied colonial rule. Nyiginya – Aristocratic lineage in Rwanda associated with the precolonial monarchical state and indirect colonial rule. Parti pour la libération du peuple Hutu (PALIPEHUTU) – Political group formed principally in the Mishamo camp in Tanzania in 1980 by Burundian Hutu refugees exiled by the 1972 violence, led by Rémi Gahutu. Parti Démocratique Chrétien PDC – Political party active in independence era in Burundi. Parti du Mouvement de l'Emancipation Hutu PARMEHUTU –Hutu nationalist party led by Grégoire Kayibanda and formed from the Mouvement Social Muhutu, active in the independence and post-independence period in Rwanda. Parti du Peuple – Populist Hutu party, strongly supported by the Belgian colonial administration, active in independence-era Burundi. Radio Télévision Libre des Mille Collines (RTLM) – Hate radio station in Rwanda which became infamous for its role in the 1994 genocide. Rwandan Patriotic Army (RPA) – Armed wing of the Rwandan Patriotic Front (RPF). Rwandan Patriotic Front (RPF) – Ruling political party in Rwanda since the 1994 genocide. Formed in southern Uganda by exiled Tutsi people. Tutsi-Banyabururi – Tutsi people from the southern province of Bururi in Burundi. Tutsi-Banyaruguru – Tutsi people dispersed throughout Burundi, but of higher social. standing due to their links to the monarchy in Muramvya province. Tutsi-Hima – Tutsi people of relatively lower social status in Burundi. Ubwiru – Codes and rituals of the Nyiginya court in Rwanda. Ubwoko – Clans of Rwanda. Ubugabire – Patron-client relationship in precolonial Burundi and colonial Ruanda-Urundi. Ubuhake – Patron-client relationship in precolonial Rwanda and colonial Ruanda-Urundi. Uburetwa – Corvée system in precolonial Rwanda and colonial Ruanda-Urundi. viii

Umuganda – Historical practice of communal labour in Rwanda. Umuganura – Precolonial sorghum festival in Rwanda. Umuganuro – Precolonial sorghum festival in Burundi. Umunyabutaka – ‘Chiefs of the land’ in Rwanda. Umunyankenke – ‘Chiefs of the long grass’ in Rwanda. Union Nationale des Etudiants du Burundi (UNEBA) – National union of students in Burundi. Union Nationale Rwandaise (UNAR) – Political party associated with the Nyiginya court active in the independence period in Rwanda. Union pour le Progrès National (UPRONA) – Nationalist party active in independence and post- independence era Burundi, led initially by Bezi prince Louis Rwagasore.

What is important to me is to understand. For me, writing is a matter of seeking this understanding, part of the process of understanding … Do I myself imagine myself being influential? No. I want to understand. And if others understand – in the same sense that I have understood – that gives me a sense of satisfaction, like feeling at home. Hannah Arendt, 1994:3 Dealing with African societies’ ‘historicity’ requires more than simply giving an account of what occurs on the continent itself at the interface between the working of internal forces and the working of international actors. It also presupposes a critical delving into Western history and the theories that claim to interpret it. Achille Mbembe, 2001:9

PART ONE

CHAPTER ONE – INTRODUCTION

Costa, 2010; Go, 2013).

Central to the analysis is a reconsideration of the relationship between modernity and tradition. It presents a critique of both the idea that modernity entails a progressive

‘detraditionalisation’ or destruction of traditional societal forms, and the idea that tradition provides a repository of cultural resources which can be used to articulate distinct and plural

‘modernities’. In the case of Burundi and Rwanda, it is argued that colonial modernity, in its indirect rule format, in important respects solidified tradition in racial terms. In the transition to independence, societal self-understandings and political movements were both enabled and 2

constrained by the legacy of colonial modernity. In the post-independence1 period, the tension between the modern commitment to autonomy and traditional legacies has been realised in profoundly destructive and violent ways. I conclude that the historical experiences of extremely violent social conflict in Burundi and Rwanda are at the same time experiences of a specific route to and through modernity.

ORIGINAL CONTRIBUTION

Put simply, the thesis is original inasmuch as it theorises the multiplicity of modernity in relation to a new substantive case which necessitates confronting experiences of indirect colonial rule and genocidal violence. As such, the analysis offers an original substantive contribution to ongoing debates at the intersection of social theory and historical sociology concerning historically and culturally differentiated experiences and interpretations of modernity. These debates are best represented in the ‘multiple modernities’ paradigm, pioneered by the Israeli sociologist S. N. Eisenstadt (2000, 2001, 2003).

This paradigm undoubtedly represents a step forward from earlier evolutionistic and

Eurocentric models of the historical development of societies. As Delanty and Mota (2015:39) state, however, there has been ‘virtually no application’ of this paradigm ‘to specific case studies within the countries of the South’. The authors’ own application to Brazil, as well as

Peter Wagner’s studies of South Africa (2012:ch. 7; 2015) are important contributions, not only in terms of extending the paradigm to excluded and marginalised cases but also in confronting the problems of colonial-imperialism and racialisation highlighted by postcolonial sociologists

1 Throughout the thesis, I distinguish between the terms post-colonial and post-independence. I use the term post-independence when I refer to the historical period succeeding the formal event of decolonisation in Burundi and Rwanda in 1962. Post-colonial, by contrast, refers to the liminal condition that emerged after the event of formal independence and before the implementation of something genuinely new, referring to both a break from colonialism and its persistence after the fact (Mbembe, 2001; Steinmetz, 2014:81; Sayyid, 2015:80-81). 3

(e.g. Magubane, 2005; Bhambra, 2007, 2014; Rodríguez, Boatcă and Costa, 2010; Go, 2013).

However, there have been no attempts to incorporate experiences of colonial-imperialism which did not involve the large-scale migration of European settlers, such as the indirect rule experienced in Burundi and Rwanda. A consideration of these countries also necessitates confronting post-colonial crises, including genocide. Eisenstadt (2000:25-26) himself suggested that Burundi and Rwanda represented ‘traumas of modernity’ but left this claim intriguingly unexplored. My thesis elaborates this claim.

As a result of this original elaboration, both its application to a novel area and its consideration of the experiences of colonial-imperialism and genocide, the thesis identifies new limitations in the ‘multiple modernities’ paradigm. It also points to where extensions of the framework are necessary so that it is more inclusive of a variety of trajectories and experiences of modernity and their ‘entanglements’ (Therborn, 2003). In this sense, this project contributes to continuing endeavours to mitigate the Eurocentric tendency to assert that the Western experience and trajectory of modernity possesses universal significance (Eisenstadt, 2000:1;

Chakrabarty, 2000:20; Seth, 2009; Bhambra, 2007; Harrington, 2016:38). It thus forms an original contribution to a much larger project of ‘global sociology’ involving the integration of social theory and regional studies whose aim is to ‘retrieve, modify and extend basic concepts of Eurocentric social theory in the light of distinctive historical experiences of other world regions’ (Arjomand, 2014:3; see also Smith, 2015:569; Ascione and Chambers, 2016).

The thesis also makes two secondary contributions due to the nature of the historical processes and events under investigation. Firstly, it makes an original contribution to the interdisciplinary field of genocide studies, as a sociological study of the relationship between genocide and modernity. Genocide studies remains influenced by an overly chronological or institutional conception of modernity that has been critiqued for its Eurocentric and evolutionist assumptions. I argue that the understanding of ‘modern genocide’ can be significantly 4

advanced by an engagement with the works of those who stress the multiplicity of modernity, its constitutive tensions, its distinct conceptions of time and the role of human agency, and the interpretive creativity of modern subjects (e.g. Therborn, 1995, 2003, 2011; Joas, 1996;

Eisenstadt, 2000, 2003; Arnason, 2003, 2003a; Bhambra, 2007, 2014; Wagner, 2008, 2012,

Delanty, 2013).

Secondly, both holistically and specifically in chapter four, the project contributes to epistemological and methodological debates about the capacity of sociology (with its Euro-

American institutional roots) to address non-Western cultures and historical experiences. I contend that a ‘global sociology’ (Therborn, 2003, 2011; Spohn, 2011; Bhambra, 2014;

Ascione and Chambers, 2016) is best grounded in a hermeneutic philosophy of social science

(Gadamer, 2004 [1975]; 2008 [1976]; Bauman, 1978; Ricoeur, 1980, 1991; Bernstein, 1983;

Harrington, 2001, 2016, Outhwaite, 2007, 2014). The critique of Eurocentrism, I argue, ought to proceed not through the wholesale deconstruction and rejection of historical concepts in the sociological tradition but instead through their application and extension, in interdisciplinary dialogue with area and regional studies (Arjomand, 2014).

AUDIENCE

This is ultimately a work in applied social theory. I seek to understand and elucidate how a set of related events and processes, that have hitherto tended to remain neglected or defined as specialist pursuits outsourced to isolated sub-disciplines, might aid the retrieval or expansion of a central category in sociology (modernity) in the light of its ongoing deconstruction and amid continuing calls for a global sociology2 (e.g. Moore, 1966; Burawoy,

2 I say ‘continuing’ on the basis that programmatic statements about globality are not especially new. See Harrington’s (2016) survey of ideas of relativity and contingency in thinkers in the interwar period in Weimar Germany, many of which prefigure notions of globalisation, reflexivity and cultural differentiation as employed today; Kurasawa’s (2004) tracing of the ‘ethnological imagination’ in the 5

2008; Bhambra, 2014; Ascione and Chambers, 2016). Area studies scholars, specialist historians and Africanists are not likely to find anything new in a strictly empirical sense in the present study. Nevertheless, I anticipate that there will be some interest in this endeavour to situate the empirical material in a broader perspective, in a way that eschews the longstanding

Eurocentric tendency to approach the African continent as a ‘object apart from the world’

(Mbembe and Nuttall, 2004:348). I also hope, as Max Weber (2001 [1905]:xl) did in appealing to ‘the Sinologist, the Indologist, the Semitist, or the Egyptologist’ in the preface of The

Protestant Ethic and the Spirit of Capitalism, that there is ‘nothing definitely wrong in points that are essential’3.

WHY BURUNDI AND RWANDA?

At this stage, I would like to justify the focus on Burundi and Rwanda, two small countries in the middle of the Great Lakes region of Central and East Africa, as case studies. I will approach this initially in autobiographical terms. If this seems indulgent now, I hope its purpose will become clearer in my discussion of philosophical hermeneutics in chapter four.

For now, it is worth remembering C. Wright Mills’ (2000 [1959]:6) famous suggestion that the sociological imagination consists precisely in the capacity to ‘grasp biography and history and the relations between the two’, an observation applicable to the orientation of the sociologist to their subject matter.

My interest in this area emerged when, at the age of 14, I saw a report on BBC News commemorating the 1994 Rwandan genocide on its tenth anniversary. It was suffused with the

history of European social thought, from Montaigne to Foucault; and Bryan Turner’s (2006) defence of classical sociologists such as Emile Durkheim against accusations of methodological nationalism. 3 He also touches upon some other pertinent methodological problems which will be addressed in chapter four (see pages 98-105): ‘It is quite evident that anyone who is forced to rely on translations, and furthermore on the use and evaluation of monumental, documentary, or literary sources, has to rely himself on a specialist literature which is often highly controversial, and the merits of which he is unable to judge accurately. Such a writer must make modest claims for the value of his work’ (Weber, 2001 [1905]:xl). 6

symbols and language typical of Western representations of Africa – bathetic lessons about good and evil, suffering, and unimaginable violence. I saw for the first time the familiar images of fractured skulls still on display at the Bisesero genocide memorial. I was struck, I recall, by two questions. One: how could this happen? And two: why did I not know about it before?

These questions remained with me in a more or less subconscious sense and only now, in retrospect, am I sure that this explains an inscrutable interest in African affairs that was never really satisfied by anything offered on my secondary school history curriculum. Instead, this interest was developed only in brief glimpses, for instance in a section on Rwanda in the

Holocaust memorial at the London Imperial War Museum on a school trip and in films, such as Shooting Dogs (Caton-Jones, 2006) about the massacre at the Ecole Technique Officielle in the Rwandan capital, Kigali.

At the University of Portsmouth, I wrote my undergraduate media dissertation on

Western images of Africa and the genocidal propaganda in the 1994 Rwandan genocide. My supervisor introduced me to Zygmunt Bauman’s (1989) Modernity and the Holocaust and with it the discipline of sociology. I was put in touch with an affiliate member of the Bauman

Institute at the University of Leeds, and after graduating, I moved north to undertake an MA is

Social and Political Thought.

My MA dissertation concerned conflict minerals in eastern Congo, their proliferation in digital technologies and consumer products, and their historical relationship to the barbarism of the Congo Free State. I was struck by my own entanglement in this history. When reading about how the war in the east of the country had intensified in 1999 in line with the demand for Colten in the production of the PlayStation 2 (Vlassenroot and Romkema, 2002:8; Mantz,

2008:42), I remembered how I had been given one to share with my brother and sisters at

Christmas. I became more and more aware of the unsettling notion that the comfortable life I 7

had led in the UK, and many things around me, were entangled with the historical violence of colonial-imperialism and its continuing and evolved relationships of exploitation and appropriation. In short, I became fascinated by how African history occupies both a marginal and fundamental position in the history of global modernity, and how areas seemingly distant in time and space are entangled in the here and now.

It is this latter theme that I am currently exploring in relation to Burundi and Rwanda.

Ultimately, the more I understood about modernity – and revealed in the process what I did not understand – the more I felt that places like Burundi and Rwanda, places so often presented as the non-modern mirror in which modern societies reflect themselves, occupy distinct and specific positions within the history of modernity rather than outside of it.

I was already familiar with Rwanda from my earlier studies but I was also keen to engage with the history of Burundi, which is too often simply equated with that of Rwanda rather than compared to it, if not ignored altogether. When I visited both countries, as well as

Uganda, prior to beginning the study I noticed sharp differences in terms of the remembrance of conflict as visible to the tourist. It was impossible to ignore the roadside shrines and memorials to victims of the 1994 genocide on bus trips in Rwanda but similar post- independence mass conflict and genocide in Burundi was barely visible at all. This reflects not only two different national approaches to post-conflict reconstruction and memorialisation, but a fundamental contrast between one case highly visible on the global plane and the other forgotten (Lemarchand, 2011; Nimubura, 2014).

However, the area studies literature reveals the intertwined constitution of the historical trajectories of Burundi and Rwanda, especially in relation to genocidal violence. It seemed necessary, therefore, from both a normative and substantive perspective, to engage with

Burundi in addition to Rwanda and to delineate how their histories converge and diverge at 8

distinct points. Often in comparative studies of genocide with social theoretical intent, history is treated in overly determinant ways, sketched over quickly as conduits to the substantive object of genocide. In Rwanda’s case (Burundi, for the aforementioned reasons, is very rarely even considered), history is effectively ‘written backwards’, seen through the prism of genocidal violence (Newbury & Newbury: 2000:832-833; Longman, 2004). A detailed consideration of the area studies literature reveals that working through the themes of colonialism and independence is equally important for understanding the experience of modernity in Burundi and Rwanda. Such a consideration demonstrates that there was nothing inevitable about the genocidal violence of the post-independence period, but neither was it purely contingent or spontaneous. The experience of colonial rule and of decolonisation – as well as precolonial dynamics – established a ‘path dependent’ 4 context in which post- independence events ought to be situated.

With this in mind, I address the development of institutional arrangements and of orientations to the world in chapter five, as well as the socio-political development of the precolonial kingdoms of Burundi and Rwanda. In chapter six, the colonial period is explored to probe questions about the relationship of modernity to tradition and thus some distinctive features of colonial modernity in its indirect form. Chapter seven focuses on the period of the formation of political parties and transitions to independence and is concerned with the diverse, conflictual interpretations of the ‘modern social imaginary’ and the emergence of competing

‘societal self-understandings’ (Taylor, 2004; Wagner, 2008, 2009) in the context of the constraints of the colonial legacy which survived formal independence and in transnational institutions. Finally, in chapter eight the Burundian genocide of 1972 and the Rwandan genocide of 1994 are explored as ‘traumas of modernity’ (Eisenstadt, 2000:25-26, 2005a). I

4 I use ‘path dependency’ in the sense implied by James Mahoney (2000:507) whereby it ‘characterises specifically those historical sequences in which contingent events set into motion institutional patterns or event chains that have deterministic properties’ (see also Skocpol, 1984:1-2). 9

will provide more detailed summaries of the chapters shortly. First, I want to present the fundamental research questions underpinning this project.

RESEARCH QUESTIONS

These questions have been formulated on the basis of the themes introduced above and encapsulate the theoretical, substantive and methodological orientations of the project. To summarise: this project contributes a substantive original contribution to the study of the multiplicity of modernity, with reference to the histories of Burundi and Rwanda, cases yet to be explored in this field; it presents a theorisation of the multiplicity of modernity and its entanglements with colonialism and genocide; and it offers a methodological contribution to contemporary debates about global sociology.

1. In the light of recent developments in the social theory of modernity, grouped together under the heading ‘the multiplicity of modernity’, what are the relationships between modernity, colonialism and genocide? 2. How do the case studies of Burundi and Rwanda offer a specific articulation of this relationship and in doing so problematise the concept of modernity? 3. What significance can be drawn from this regarding the capacity of sociology and its Western roots to engage with historical developments and events in the non-Western world?

STRUCTURE OF THE THESIS

The thesis is presented in three parts. The first part is generally expository. It includes an engagement with the relevant literature as a means to situate the work in debates in contemporary social theory and genocide studies, an elucidation and definition of key terms, and a development of the theoretical framework and methods to be utilised in the pursuit of the research questions. Woven throughout part one is a consideration of how the theoretical and methodological debates discussed come to bear specifically on the cases of Burundi and

Rwanda. This serves a double function. On the one hand, it serves to avoid a ‘top-heavy’ 10

account which separates the general theoretical elements of the thesis from particular idiographic detail to a degree whereby their connections are obscured. On the other, it serves as an extended introduction to elements of the case study which will be explored in greater depth in part two.

The second part of the thesis is a detailed empirical case study of Burundi and Rwanda, functioning as a means to explore in-depth the themes and problems raised in part one. Each of these chapters is centred on a distinct theoretical problem. These problems are introduced at the beginnings of the chapters and frame substantive-empirical analyses. They are revisited at the end of each chapter in the form of an intermediate reflection which situates their conclusions in relation to broader parameters of the project.

I adopt a comparative focus within discrete chapters, rather than presenting Burundi and Rwanda’s trajectories in separate sections before a comparative-analytical chapter. Firstly, this is a clearer way of organising the thematic and empirical material for a reader presumed intelligent though unfamiliar with the cases. Secondly, this organisation allows more detailed comparisons of Burundi and Rwanda, and perhaps more importantly to elucidate to a better extent the entanglement of their historical trajectories to and through modernity. The result is that each of these chapters can be read as self-contained essays, but together form indispensable

‘parts’ of the ‘whole’ of this thesis, to use hermeneutic terminology.

The thesis is concluded in part three. Here, I address the research questions outlined above and emphasise the original contribution that the thesis makes, and I also outline a future research agenda. Finally, I point to some of the normative implications that arise from my analysis in relation to more contemporary events, in the Great Lakes region and beyond.

CHAPTER SYNOPSIS

PART ONE In chapter two, I outline some conceptual, definitional and normative problems to do with the identification of genocide and colonialism and their interrelations for the purposes of this study. In doing so, I engage principally with sociological work in genocide studies and in postcolonial sociology, situating my own work in relation to them as well as defining them for the purposes of ascertaining their connections to the theorisation of modernity which follows.

In chapter three, I present the central historical and theoretical framework within which the substantive analysis will be situated; that of ‘entangled historical routes to and through modernity’ (Therborn, 2003). In doing so, I distinguish modernity from associated concepts of

‘modern society’ and ‘modernisation’. I argue that modernity is best conceived as a time- orientation, whereby the future is conceived of as a space for projecting possibilities within the present. These possibilities are situated within a field of tensions and require active, creative and often conflicting interpretation, which give rise to diverse institutional arrangements and social practices. These are formed against the background of distinct historical experiences (for instance, of colonialism, or of revolution) and socio-cultural traditions and unfold processually along specific historical trajectories or routes. I subsequently define modernity as a constitutively plural phenomenon articulated in a variety of specific forms which are entangled both with one another and with transnational institutions.

In chapter four, I suggest that philosophical hermeneutics can provide a thorough basis for this historical sociological project in the philosophy of social science. As such, it contributes to some contemporary programmatic calls to integrate social theoretical concepts with the findings of area studies in order to account for non-Western historical trajectories which have historically been excluded from the Eurocentric historical narration of the emergence of modernity (Arjomand, 2014; Bhambra, 2014) I will then move on to a more practical 12

discussion of how these methodological issues impact this particular study, before outlining the methods employed specifically in the case study chapters which form part two.

PART TWO Chapter five explores the precolonial histories of the monarchical states of Burundi and

Rwanda in comparative perspective. Firstly, I situate them within the wider Great Lakes region, an entity culturally and ecologically distinct from the rest of sub-Saharan Africa, and outline some commonalities in their cultural programmes and institutional forms. I then delineate some important divergences in the precolonial developments of Burundi and Rwanda, specifically concerning the social identities of ‘Hutu’ and ‘Tutsi’, which have to be factored in when considering developments in the colonial and post-independence periods.

Chapter six is a study of the colonial ‘entanglement’5 of Burundi and Rwanda, oriented towards social- and political-theoretical questions about modernity. Overall, I identify a number of policies and actions, particularly in reforms made by the Belgian colonial administration from 1926-1936, that ‘solidified’ various elements of the precolonial societies of Burundi and Rwanda, administered in unison as Ruanda-Urundi and attached to the larger territory of the Belgian Congo. Far from destroying traditional societies in steam-roller-like fashion, indirect rule in some ways worked ultimately to preserve tradition and harness the non-teleological dynamism of precolonial societies. Although the colonial governments also introduced new dynamics associated with Western modernity – such as market capitalism,

Christian morality, education systems – it is not the case that ‘all that was solid melted into air’

(Marx and Engels, 2004 [1848]:7). This is particularly apparent in the streamlining of various precolonial social institutions along overtly racial lines. I suggest that this institutional feature of 19th Century indirect colonial rule – which I suggest might be called a colonialisation of the

5 See pages 72-77 for a justification for this term. 13

‘cultural programmes’ of colonial societies – is an unduly underemphasised theme in the paradigm of multiple modernities.

Chapter seven is an examination of the transitions to independence of Burundi and

Rwanda during the late 1950s and early 1960s. Although Burundi and Rwanda were entangled administratively as Ruanda-Urundi, Rwanda became independent as a republic via a revolution whereas Burundi became a constitutional monarchy by way of a national liberation movement.

Here, I am concerned with the competing interpretations among elites of the social imaginary of modernity in Burundi and Rwanda. I explore how these interpretations are both enabled and constrained by the experience of modernity imposed by indirect colonial rule, as well as by global transformations. In short, a space opened up for a diverse variety of ‘societal self- understandings’ (Wagner, 2010a), the possibilities therein projected into a future shapeable in autonomous terms. Also implied in these self-understandings were distinct narrative relationships to historical traditions and events and precolonial social formations, which had been racialised during the experience of indirect rule. I look specifically at the elite social discourses emanating from political parties in each country and present their trajectories to independence in comparative perspective.

Chapter eight unpacks Eisenstadt’s (2000:25) intriguing but under-elaborated claim that postcolonial Burundi and Rwanda comprise a ‘trauma of modernity’, of the sort that indelibly marked the twentieth century and ‘brought into question its great promises’. Here, the periodic destruction and genocide that occurred in the entangled post-independence experiences of Burundi and Rwanda were ‘not outbursts of old ‘traditional’ forces, but the result of the ongoing dialogue between modern reconstruction and seemingly traditional forces’

(Eisenstadt, 2000:25-26). The genocides in Rwanda 1994 and Burundi 1972 will be situated within the context of their entangled historical routes to and through modernity. Post- independence Burundi and Rwanda demonstrate to a tragic extent that experiences and 14

interpretations of modernity can give rise to violent – indeed, genocidal – social conflict and trauma.

PART THREE In the conclusion presented in chapter nine, I gather together the responses to the research questions for chapters five to eight and orient them towards the theoretical and methodological contributions of the thesis as a whole. I then address the recent history and contemporary situation of the Great Lakes region not addressed within the remit of this thesis, outlining in particular some policy implications in terms of atrocity prevention. I also present some more speculative points on global transformations and their implications for the relationship between modernity and genocide. Finally, I point to areas for further development and inclusion in a future research agenda.

CHAPTER TWO – GENOCIDE AND COLONIALISM: PRELIMINARY CONCEPTUAL AND DEFINITIONAL PROBLEMS

At this point, it is necessary to elucidate what is meant by two key terms to be employed throughout this thesis: genocide and colonialism. It is also necessary to address some conceptual and normative problems that empirical studies of genocide and colonialism must necessarily confront. I will treat genocide and colonialism in turn, before addressing them together in a discussion of recent historical scholarship in genocide studies. This will pave the way for their incorporation into the theorisation of the concept of the multiplicity of modernity to follow in chapter three.

GENOCIDE

MORAL, POLITICAL AND LEGAL DIMENSIONS OF GENOCIDE A consideration of the literature in genocide studies and broader debates concerning violence in contemporary societies immediately raises definitional difficulties. The first problem arises when trying to distinguish genocide as an observable event from a moral- historical concept. Though its moral and empirical components may be separable analytically, in practice the definition of genocide is not simply a neutral designation of a factual category.

It is a moral act, often situated in highly-charged political contexts. To declare genocide is to declare evil.

The definition most commonly employed is that of the Convention on the Prevention and Punishment of the Crime of Genocide (1948) which developed from the work of the legal theorist, Raphael Lemkin. Here:

Genocide means any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such: (a) Killing members of the group; (b) Causing serious bodily or mental harm to members of the group; (c) Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; (d) Imposing measures intended to prevent births within the group; (e) Forcibly transferring children of the group to another group.

As a legal instrument, the convention has frequently been ineffective since its inception, as have later developments such as the concept of ‘crimes against humanity’, which was designed to broaden the remit of international law in response to a range of atrocities not captured under the above definition of genocide. Institutions such as the International Criminal

Court (ICC), established in 2002 following the Rome Conference, have repeatedly had difficulty in successfully prosecuting accused war criminals such as President Omar al-Bashir of Sudan, who was charged in relation to the Darfur atrocities in 2009. The ICC has not even been able to issue warrants in the case of systematic human rights violations by the governments of Sri Lanka, Israel or China, let alone for possible war crimes committed in Iraq by US and UK forces (for discussion see Feierstein, 2014:11-38)

While a legal definition might conceivably help us to interpret culpability for genocide or to demonstrate that genocide has taken place, it does not explain or aid understanding of why genocide occurs. Furthermore, elements of the definition are problematic from a sociological perspective, such as the nature and definition of ‘intent’. There are also controversies surrounding the notion of ‘groups’, both in terms of how groups are socially constituted and the omission of important categories. Political groups, as distinct from national, ethnic, or religious groups, were excluded from the convention, even though they were central to

Lemkin’s earlier sociological definition and are frequent targets of state violence6 (Feierstein,

2014:29-31).

6 Martin Shaw (2007) presents an extended sociological critique of how the convention constitutes a narrow legalistic interpretation of Raphael Lemkin’s broader sociological definition. See the following from Lemkin (2009 [1945]:6): ‘The term [genocide] does not necessarily signify mass killings although it may mean that. More often it refers to a coordinated plan aimed at destruction of the essential foundations of the life of national groups so that these groups wither and die like plants that have suffered a blight. The end may be accomplished by the forced disintegration of political and social institutions, of the culture of the people, of their language, their national feelings and their religion. It may be accomplished by wiping out all basis of personal security, liberty, health, and dignity. When these means fail the machine gun can always be utilised as a last resort’.

Furthermore, the convention definition de-emphasises genocidal expulsions, now referred to as ‘ethnic cleansing’. Lemkin’s definition was intimately concerned with space: genocide meant the destruction of a particular group within a given territory. However, violent population removal was not incorporated into the convention (Schabas, 2000:196). This is largely attributable to the context of realpolitik at the time when the convention was being drafted and ratified. In the aftermath of the Second World War, for instance, millions of ethnic

Germans were forcibly removed from Central and Eastern Europe, and hundreds of thousands of people were expelled from Palestine during the establishment of the state of Israel, each under the auspices of prominent international powers (Shaw, 2013:78-89).

The mass killing of populations is a perennial aspect of human history (Eisenstadt,

2005a:635; Bloxham and Moses, 2010:ch. 12-16). We might, however, say that the concept of genocide is a phenomenon of modernity in the sense that the use of the term gave a name to a crime perpetrated against humanity tout court, a crime that could in principle be prevented or even eradicated in a future shapeable by human agency in the name of peaceable progress.

Paradoxically, however, genocide has persisted in the modern period. It has even been convincingly argued to have been facilitated by the institutional features of modern societies and has been committed with a future-orientation, in the names of progress, revolution and nation (Bauman, 1989; Stone, 1999, 2004a; Eisenstadt, 2000, 2005, 2005a; Levene, 2005,

2005a; Mann, 2005; Bloxham, 2008; Moses, 2008; Shaw, 2013).

This tension is a central problem within genocide studies, a field which began with single case studies of individual genocides modelled on the study of the Shoah7. By the 1980s, a small group of scholars had began to engage in comparative analysis of multiple cases of

7 Shoah is used as distinct from the Holocaust because it refers to the distinctly Jewish dimension. The Holocaust is used to refer to the collection of genocidal campaigns waged by the Nazis, including the destruction of Slavic peoples, Romani and Sinti people, gay people, people with mental illness, and people with physical impairments. 18

genocide (Kuper, 1982; Chalk and Jonassohn, 1990; Fein, 1993; Hiebert, 2008:309). The field has continued to grow and specialist journals began to appear in the 1990s, flourishing amidst the international reactions to genocide in Rwanda and the Balkans and the resulting rise to prominence of an increased global concern with human rights and atrocity prevention. Recent years have seen the appearance of a number of readers, handbooks and introductory texts

(Totten and Bartrop, 2009; Bloxham and Moses, 2010; Jones, 2011), as well as an avowedly

‘critical’ turn (Moses, 2008b; Hinton, 2012). In practical terms, genocide prevention is now the raison d’être of various government agencies, NGOs, and specialist units at the United

Nations (Hinton, 2012:4).

Nevertheless, genocide studies has since its inception been marked by a

Historikerstreit8, centred on how events other than the Shoah ought to be situated. For some, trying to understand the Shoah in comparative terms with other genocides is an unacceptable denial of its unique evil (Fackenheim, 1978:ix; Bauer, 2001:10-12; Katz, 2001:49). For many more, it is implicitly or explicitly posited as the ideal typical case of genocide against which others’ genocidality should be measured (for critique Powell 2011:88-89). Even for Bauman – who heavily criticised the presentation of the Holocaust as a unique and particular event in

Jewish history which makes it ‘comfortably uncharacteristic and sociologically inconsequential’ and belittles its significance for ‘sociology as the theory of civilisation, of modernity, of modern civilisation’ (1989:1) – the Shoah is uniquely modern, and the modern firmly equated with the West9.

8 I refer to the Historikerstreit or ‘historians’ debate’ which played out in the late 1980s and concerned the memorialisation of Nazi Germany and the legacy and contested comparability of the Holocaust (see Varcoe, 1998:61). 9 This may explain Bauman’s (2008:103-109) lack of engagement with genocide in Rwanda in 1994. There is, however, an intriguing engagement with Stalinism in Bauman’s oeuvre (Bauman, 1985, 2004; Beilharz, 2002), which might complement analyses of Soviet trajectories and interpretations of modernity (see Arnason, 1993, 2000). With this critique in mind, I proposed an extension of Bauman’s ideas towards a consideration of European colonial relationships at the International Social Theory Consortium meeting in 2015 (Palmer, 2015). 19

This has resulted in a situation whereby a large proportion of work has been produced with the political aim of recognition, of ascertaining whether or not a particular event or set of events constitutes genocide (Shaw, 2013:25). In this sense, genocide studies reflects transnational cultural debates about how traumatic events experienced by various collectives, such as the histories of Atlantic slavery and European colonial-imperialism, ought to be memorialised vis-à-vis the Holocaust. It is thus connected to a much broader context of communal struggles for the recognition and memorialisation of historical suffering, and what is sometimes disparagingly called ‘identity politics’ (Fraser, 2000; Wieviorka, 2012:8; Kern et al, 2014).

It also draws attention to the paradoxical relationship between the universalising and particularising dynamics of what Jeffrey Alexander (2009, 2012) terms ‘cultural trauma’. For

Alexander, the memorialisation of the Holocaust has a universalising thrust, in that it involves a process of making a specific historical event affecting a particular community a model for much broader processes of ethical commemoration and reconstruction. Notwithstanding critiques of the spectatorial voyeurism of what some see as the assimilation of the Holocaust into the ‘culture industry’ (Finkelstein, 2000; Adorno, 2001; Wood, 2010), this raises the question of how to understand those fiercely particularistic forms of collective memory which feed into narratives of exceptionalism and material practices of securitisation and pre-emptive defence. One of the paradoxes of the legacy of the Twentieth century, which has been called the ‘century of genocide’ (Totten et al, 2009; Weitz, 2005), is that the declaration of ‘Never

Again’ which emerged alongside the ratification of the 1948 genocide convention has been invoked just as often as a call to arms as it has been a call for preventative awareness10.

10 A counterpoint to the universalising orientations of the cultural trauma framework is presented in Dirk Moses’ proposition of ‘the terror of history’, borrowed from Mircea Eliade (1959) and invoked to argue that ‘when current events are depicted as reincarnations or perpetuations of the traumatic, often genocidal, experience, the ensuing posture is defensive, leading to preemptive or anticipatory self- defence’ (Moses, 2011:96-97). 20

Genocidal violence in Burundi and Rwanda is a case in point. As I shall demonstrate in chapter eight11, it cannot be sufficiently understood without consideration of perpetrators’ perceptions of historical injustices and experiences of violence, shared by groups who construct such events as evil. Some Burundian Hutu refugees in Rwanda were participants in the genocide perpetrated against Tutsi in Rwanda in 1994 (Des Forges, 1995:197). These refugees had fled inter-ethnic massacres in the wake of the 1993 Tutsi-led coup and murder of Melchior

Ndadaye, the first (Hutu) democratically elected president of Burundi, itself perceived as a replay of the 1972 genocide against Hutu (Malkki, 1995; Lemarchand, 2009:75).

Furthermore, the Tutsi Rwandan Patriotic Front (RPF), the current governing force of

Rwanda whose seizure of power stopped the 1994 genocide, has been accused of perpetrating genocidal atrocities in eastern Congo – in the name of securitising Hutu refugee camps – in the years after the genocide (United Nations High Commissioner…, 2010; Lemarchand and

Reyntjens, 2011; Okosan and Kibiswa, 2013). The RPF themselves invaded Rwanda in 1990 from Southern Uganda, where they had been in exile since the anti-Tutsi violence which accompanied the 1959 Rwandan social revolution, beginning the civil war in which the 1994 genocide has to be situated.

The role of forced migratory patterns and refugees as carriers and subjects of cultural trauma in the region thus compels us to move beyond narrow interpretations focussed on spontaneous eruptions of violence contained within nation states. In Burundi and Rwanda, genocide and ethnic violence have often been the culmination of a rational, if ‘delusional’

(Semelin, 2007), security decision made by the perpetrators, and the subsequent mobilisation of a retributive response by a fearful section of a population against a perceived external threat, a threat which has often been quite real. The targets are defined as both outside and within national borders. Genocide is connected to the logic of counter-insurgency.

11 See pages 214-238. 21

Clearly, then, there are problems of value- and political-orientation inherent in any study of genocide. I claim that it is desirable and possible, to a degree, to attempt to reflexively disentangle or detach oneself from these orientations. My concern is not with contributing a neglected case to a taxonomy of violence or with furthering specific causes in competitive terms with better memorialised cases, worthwhile though these pursuits may be. Rather, mine is an original contribution to an understanding of the various and specific ways in which genocidal violence was experienced throughout the nineteenth and twentieth centuries, situated in broader experiences of entangled routes to and through modernity. I propose that this constitutes a way of doing justice to the ‘specificity’ (Hallwood, 2001) of genocide and genocidal violence in Burundi and Rwanda. They are not unique, singular or incomparable instances and neither are they reducible to nomothetic historical laws; rather they are specific events entangled in various broader historical processes of racism, colonial-imperialism and nation-building (Moses, 2002:28; Stone, 2004:127; Lentin, 2008:494). There is a strong affinity, I suggest, between the emphasis on the multiplicity of modernity and the practice of what has been called ‘multidirectional memory’ (Rothberg, 2009; Silverman, 2013).

DISTANCING, PROCESS REDUCTION AND CONCEPT REDUCTION We still have to arrive at a workable definition of genocide. In order to do so, I will present a brief critique of two tendencies which I label ‘process reduction’ and ‘concept reduction’. Process reduction, a concept formulated by Norbert Elias (1978:113-116), refers to the reification of social processes to the extent that they take on the status of a static, isolated object in a state of rest. Here, it is encapsulated by the idea that genocide is solely an ‘event’, detached from the processes that precede it, and thus appears as an extraordinary action well outside the realm of ‘ordinary’ social practices. Concept reduction, a term which I borrow from 22

the philosophy of science (Bunge and Mahner, 1997), refers to the notion that genocide is identical with the processes from which it emerges12.

My main critique in terms of process reduction is aimed at a prominent conception of genocide which tends to be associated with stigmatised populations (Moses, 2008b) construed as ‘distant’ from the West in various ways (see Kressel, 1996; Power, 2002; Tatum, 2010;

Stanton, 2013). In spatial terms, this distance is produced by the assumption that genocides typically occur in faraway ‘local’ contexts, and that globality is only relevant in terms of whether developed ‘Western’ states or international organisations ought to intervene to stop these ‘local’ genocides (Meister, 2012). Genocide is presented as an evil of ‘bad’ countries – totalitarian; communist; fascist; ‘failed states’ (Harff, 2003) – rather than as an event, emergent in some social configurations from the entanglement of various processes and structures operating both within and across state boundaries. Important here, as Burundi and Rwanda demonstrate, are historical colonial relationships and their persistence in ‘coloniality’

(Maldonado-Torres, 2007; Mignolo, 2012; Ndlovu-Gatsheni, 2013) after the formal event of independence, as well as global economic institutions such as the World Bank, the power blocs of the Cold War, international political and legal institutions, NGOs and development assistance and arms trade networks.

Process reduction also distances genocide in temporal terms. Genocides are deemed to happen in places at an earlier, ‘backward’ stage of development. They are commonly referred to in terms of ‘ancient tribal hatreds’ and other descriptors that suggest timelessness. It is well established that this is a frequent trope in the Western mediation of African conflicts, of which

Burundi and Rwanda are particularly prominent examples (McNulty, 1999; Carruthers, 2004;

12 To quote Bunge and Mahner (1997:114), ‘to reduce a concept A to a concept B is to define A in terms of B, where B refers to a thing, property, or process on either the same or on a lower (or higher) level than that of the referent(s) of A’. This mirrors some arguments in realist social theory about stratified ontology and the emergent properties of social phenomena (see Sayer, 2010). 23

on the general problems of the representation and interpretation of ‘distant suffering’, see

Tester, 1997; Boltanski, 1999; Chouliaraki, 2006). What is ultimately produced is a ‘denial of coevalness’ (Fabian, 1983). The violent histories of the ‘developed’ nation-states from which these discourses emanate, particularly in the context of colonial-imperialism, are deemed to be firmly in the past, if not ignored altogether. In this sense, the effect of spatio-temporal distancing is that genocide appears as a phenomenon that is far away and long ago.

Distance is further maintained in various ways. It might be said to be produced in the very constitution of genocide studies. The separation of genocide from mainstream sociological research and theorising, its specialisation in the field of genocide studies, perhaps implies that genocide, inasmuch as it seemingly concerns the destruction of the social, is ultimately an anti- sociological phenomenon (see Bauman and Weltzer, 2002). This thesis also presents an implicit critique of a certain form of comparative historical study of genocides that attempts to construct a trans-historical and nomothetic general theory of genocides across a wide array of cases. Though they have produced some extremely valuable results to which I am indebted, these tend to take specific genocides out of their regional and historical contexts, thus distancing them from each other and the connections between them 13 . This is quite straightforward in terms of my thesis: the Rwandan genocide in 1994 is frequently treated

13 A critique of the comparative historical sociology of modernity in favour of focusing on ‘connections’ is presented by Gurminder Bhambra (2007:72, 2016; see also Magubane, 2005:95). Along similar lines, a critique of the comparative method as it pertains to genocide studies is presented by Shaw (2013:31), who argues that ‘comparative genocide studies has… got badly stuck with the idea that what it is studying are discrete ‘genocides’’, rather than the phenomenon of genocide as it is situated in the transformations of modernity. My own focus on entanglement entails a focus on divergence of national and local trajectories as well as their convergence with broader transnational and global processes. As such, I agree with Jürgen Kocka (2003) that comparison and connection are not mutually incompatible but ought to operate in tandem. 24

without reference to the aforementioned regional context, including Burundi14 (e.g. Powell,

2011:ch. 7; Sagall, 2013:ch. 8).

The effort to construct trans-historical and global theories of genocide tends to privilege a certain kind of genocidal violence that fits into pre-defined typologies. These cases are typically what Mark Levene (2005:163) calls ‘mega-genocides’, referring to frequently discussed cases like the Shoah, Armenia, Rwanda and Cambodia. More ‘contentious’ cases, such as the genocides in Burundi in 1972, Indonesia in 1965 or eastern Congo in 1996, are typically not included in these accounts, for various reasons including political expediency, conceptual fuzziness and historical amnesia (Lemarchand, 2011; Hinton et al, 2013).

The sense in which the seeking of ‘hard knowledge in the form of universal laws with predictive potential’ so that genocide can be mapped ‘in the manner of a mathematical equation’

(Moses, 2008b) also tends to engender an overly deterministic view of history which downplays both contingency and human creativity. In the words of the philosopher of creativity,

Cornelius Castoriadis (1997:4):

Determinacy leads to the negation of time, to atemporality: if something is truly determined, it is determined since always and forever. If it changes, the ways in which it can change and the forms that this change can bring about are already determined. Then ‘events’ only realize laws, and ‘history’ is but the unfolding along a fourth dimension of a ‘succession’ that, for an Absolute Mind (or for the accomplished scientific theory), would only be coexistence. Time is then sheer repetition, if not of ‘events’, then of the instantiations of laws. This is not a defeatist declaration that genocide cannot be prevented or that its processes might not be better understood and intervened in. Rather it is to say that understanding and preventing genocide, as the pragmatist philosopher Richard Bernstein (2002, 2013) says of radical evil and violence, is an ongoing hermeneutic activity that in principle can never be exhausted. To

14 Where the Great Lakes region is taken as a whole, this is generally in short chapter-length works (e.g. McDoom, 2010; Jones, 2014) which are more concerned with assessing whether certain cases can be classed as a genocide or not, i.e. with the political aim of recognition. 25

do this means to attempt to understand much broader patterns of social change and the possibilities of the emergence of genocide immanent in them.

Extermination, then, is better placed at the extreme end of a continuum including more quotidian practices of social exclusion and dehumanisation, a notion which disturbs the idea that genocide can be unproblematically distanced. Much critical theoretical work has presented compelling arguments warning of the violent potentialities in the midst of avowedly developed, modern and democratic societies (Adorno and Horkheimer, 1997 [1944], Bauman, 1989;

Agamben, 1998; Scheper-Hughes and Bourgois, 2004, Mann, 2005). Several genocide scholars, influenced by Raul Hilberg’s (1985) work on the destructive process of the Holocaust, have developed early-warning systems premised on the notion that genocide is a processual phenomenon, which passes through observable organisational stages such as classification, dehumanisation and polarisation (Rosenberg, 2012; Stanton, 2013; Feierstein, 2014:104-130).

Such an approach, however, runs the risk of reducing the concept of genocide to these practices themselves. This is what I mean by ‘concept reduction’. Such reduction is present when genocide, or associated terms such as Holocaust, becomes a discursive weapon in the sorts of memorialisation struggles outlined above15. The dehumanisation of stigmatised outgroups who are assigned pejorative characteristics or are perceived as threatening to the majority is arguably common to all societies. In the vast majority of cases, these groups are not subject to the sorts of systematic violence that might be called genocide. Genocide ought to be

15 Charges of genocide, as Helen Fein (1990:5) noted, appear in as diverse places as anti-abortion discourses, activism against the arms trade or against the ‘cleansing’ operations of urban gentrification. The dangers of such politicisation can lead to some extremely crude appropriations, for instance by white supremacists, some of them deniers of the Holocaust, who perceive that diversity or ‘mass immigration’ is a ‘genocide against the white race’. On better, though still shaky, ground are Evans and Giroux (2015:129, 137-138), who wrote that state violence and oppression of black people in Ferguson, Missouri ought to be construed as genocidal.

seen as a form of social action with emergent properties, which arises from the interaction of various sub-processes but is not sufficiently determined by them.

Various genocide scholars have sought to transcend this difficulty by differentiating between the noun ‘genocide’, designating an event, and the adjective ‘genocidal’, which distinguishes certain kinds of action. This extends to the pioneering work of the South African political sociologist, Leo Kuper (1982:10), and his distinction between genocide and

‘genocidal massacres’. Levene (2008:311) differentiates between genocide and ‘genocidal processes’, the latter involving ‘all manner of draconian or coercive measures, ranging from the forcible assimilation of a group at one end of the spectrum through to physical murder at the other’. Martin Shaw (2013:6) makes a similar distinction between genocide and ‘genocidal violence’, the latter inclusive of ‘limited-scale genocidal expulsions, genocidal mass rapes, genocidal starvation policies, etc.… targeted, destructive violence against population groups which is perpetrated episodically, locally or on a small scale, in situations where it does not seem appropriate to talk of ‘a genocide’’. Throughout this thesis, I utilise Shaw’s vocabulary.

Still, these terminological distinctions do not by themselves sufficiently deal with how genocide emerges out of these related forms of violence. It is necessary to identify the conditions of emergence of genocide. Genocide has been identified as generally occurring in times of acute crisis and upheaval. As Bauman (1989:114) noted, ‘deep social dislocations’ are the ripest spaces for modern genocide, when society is both ‘malleable’ and ‘helpless’.

Economic crises and shortages in resources like food and land have been identified as precipitating factors of genocide in Germany, Cambodia and Argentina (Spencer, 2012:49).

The collapse of the global coffee market in the late 1980s and early 1990s, combined with the profoundly deleterious effects of structural adjustment programmes overseen by the World

Bank and the International Monetary Fund (IMF), had a significant impact in terms of the 27

development of the 1994 genocide in Rwanda and wider regional strife, including in Burundi

(Chossudovsky, 1996; Kamola, 2007).

The political upheaval created by revolutions or coups has also fomented conditions for genocide in countries as diverse as Cambodia, Ethiopia and Armenia (Melson, 1992; Kissi,

2006). Levene (2005a:103-213) argues that the Jacobin crushing of the Vendée uprising signifies the emergence of modern genocide (see also Eisenstadt, 1999). In chapter seven16, I will discuss the Jacobin component of the 1959-62 social revolution in Rwanda, before looking at how postcolonial anti-Tutsi genocide has been committed in fidelity to this foundational event. Burundi’s tumultuous postcolonial political history is punctuated by coups d’état, often followed by systematic mass violence. Here, abortive Hutu-led coups in 1965, 1972 and 1988 led to genocidal violence of varying degrees of intensity. In Rwanda, genocide in 1994 was initiated following the take-down by missile of President Juvenal Habyarimana’s plane (BBC

News, 1994).

Relatedly, there has emerged in genocide studies a consensus that genocide overwhelmingly takes place under the acceleratory conditions of war (e.g. Shaw, 2003, 2007a;

Bloxham, 2007, 2009). Rwanda had been in a state of civil war for four years by the time of the 1994 genocide. As I hinted above, genocide and genocidal violence can be seen as radical forms of counter-insurgency warfare.

INTENTIONALITY AND CREATIVITY Genocide ought to be seen as a ‘situationally creative’ (Joas, 1996) form of collective action, though it is not determined by the situation out of which it emerges (inasmuch as genocide is sui generis over and above its constituent processes). It possesses an improvisatory character. Genocide, as Michael Mann (2005:7; see also Schneiderhan, 2013) argues, is rarely

16 See pages 181-191. 28

the original intention of perpetrators. It tends to emerge, beyond compromise and a range of repressive measures, as a sort of ‘Plan C’. Genocide is not the result of bursting forth of spontaneous passions. Examinations of those cases that are often referred to in order to support this idea17 reveal that there are significant ordering imperatives behind genocide. Neither, however, can genocide be seen solely as a linear process of rational planning and efficient organisation. For instance, to take Abram de Swaan’s (2015:42-43) example, ‘the Nazi

Endlösung was mostly a sequence of ad hoc and improvised measures in reaction to challenges and opportunities that the quickly moving war theatre presented’.

At this point, I need to establish what I mean by ‘creativity’. Serious care needs to be taken with the concept, which has acquired a normative and fashionable reputation alongside the emergence of neoliberal economic policies (Boltanski and Chiapello, 2007). To be clear, in no way is it my intention to be purposefully controversial or contrarian. The stakes are far too high for this. Rather, my suggestion here is that the problem of intentionality, both in theoretical debates within genocide studies and in legal and political applications of the definition of the genocide convention, necessitates a confrontation with what the neo- pragmatist sociologist Hans Joas (1996) has called ‘the creativity of action’. This implies seeing creativity in neutral or ambivalent (Bauman, 1990, 1991) terms.

There are some influential precursors in this regard. For instance, Arendt’s (1967

[1951], 1990 [1963], 1994, 1998 [1958]) concern with natality, the human capacity to ‘begin

17 There is a notable scene in Joshua Oppenheimer’s (2014) documentary, The Look of Silence, where a military general who in 1965 had overseen the killings of many suspected communists in Indonesia attempts to avoid moral responsibility for his role in the massacres by imputing them to the spontaneous passions of masses. Similarly, whilst killing was taking place in Rwanda in 1994, leading génocidaires similarly attempted to explain away the nature of the violence by reducing it to uncontrollable outbursts of savagery from the population. This explanation was regurgitated by large sections of the international news media and was a significant factor in the lack of intervention (Melvern, 2006:235). What both examples demonstrate is that recourse to explanations which stress the ‘spontaneous whims of the masses’ above all else can have the perverse effect of vindicating the perpetrators. 29

anew’, applied as much to her analyses of totalitarianism and concentration camps as novel and unprecedented phenomena as it did to the American Revolution or the Greek polis. Castoriadis

(1991:161) also held that the definition of creativity need not entail any normative evaluation:

‘creation does not necessarily – nor even generally – signify ‘good’ creation or the creation of

‘positive values’. Auschwitz and the Gulag are creations just as much as the Parthenon and the

Principia Mathematica’.

That Bauman has drawn influence from both thinkers perhaps explains the emphasis on possibility in his analysis of the Holocaust and in his sociology more broadly (Bauman, 1989;

Tester, 2004). Throughout Modernity and the Holocaust (e.g. 1989:17-18), there are frequent references to genocide as a ‘potential’ or ‘possibility’ of modernity, rather than an inevitability or probability. It is this aspect that points to the ethical significance of the book in terms of evaluating the present. In his (1989:84) words:

True, the Holocaust occurred almost half a century ago. True, its immediate results are fast receding into the past. The generation that experienced it at first hand has almost died out. But – and this is an awesome, sinister ‘but’ – these once-familiar features of our civilisation, which the Holocaust had made mysterious again, are still very much part of our life. They have not gone away. Neither has, therefore, the possibility of the Holocaust.

Beyond this, Bauman (1989, 1991:18-53, 1995:192-206) also pointed to the creative aspects of modern genocide when he highlighted its ‘order-building’ rationalisation. The genocides of Nazism and Stalinism were practices of creative destruction that would, in the minds of the perpetrators, bring about a better world in line with utopian blueprints. More recently, the Argentinian sociologist Daniel Feierstein (2014) has also identified that genocide is in some senses a creative, ‘reconstructive’ social practice, insofar as it involves a radical reconfiguration of social relations as opposed to their destruction. He defines genocidal social practices as ‘a technology of power intended to destroy social relations based on autonomy and cooperation by killing a significant portion of society (significant in numbers or influence), that then attempts to create new social relations and identity models through terror’ (Feierstein, 30

2011:262 emphasis added). As such, genocide ought to be grouped in the same class of transformational, macro-social phenomena as revolution (Feierstein, 2014:37).

The work of Joas both complements and complicates these accounts. Common to

Bauman and Feierstein is an overly rational and teleological idea of action that only really accounts for the ‘order-building’ or ‘restructuring’ orientations of regime elites. They do not sufficiently explain how people from other social strata are incorporated into genocide, as they were in large numbers in Burundi and Rwanda, beyond the assumption that they merely assimilate the values and directives of elite discourses. Nor do they emphasise the conditions which lead genocide to emerge from more quotidian forms of discrimination.

Joas does not propose ‘creative action’ as an additional category to be placed alongside instrumental-rational, value-rational, normative or habitual forms of action. Rather, he attempts to show that there is an inherent creativity common to all action (Joas, 1996:144; Joas & Knöbl,

2009:514). Three components are pivotal to this idea: situation, corporeality and sociality.

These words form part of a constructive critique of theories of rational action which tend to isolate the individual actor from the surrounding context in which they act; assume that that they are always capable of purposive action and pursuing goals teleologically; assume that they have control over their bodies; and assume that they are autonomous in relation to other actors

(Joas, 1996:146-147).

Firstly, action is always situated, which is to say that it takes place in contexts of limiting and enabling conditions. These conditions themselves affect the orientation of action:

‘Only when we recognize that certain means are available to us do we discover goals which had not occurred to us before’ (Joas, 1996:154). As such, action is a process, which in its unfolding changes course in response to the changing conditions in which it is situated. This is 31

a challenge to those teleological conceptions of action which see goals as pre-defined before they pursued in action:

It is the specific situation in which action occurs, in which processes of perception and cognition occur, in which plans and goals are formulated when new situational interpretations crop up … These situational challenges thus require new and creative solutions rather than the unwavering pursuit of goals and plans formulated at a particular point in time. Motives and plans are products of reflection within action situations, not (antecedent) causes of action (Joas & Knöbl, 2009:518; see also Joas, 1996:158). Furthermore, the body does not play an instrumental or subsidiary role in human action

(1996:168). Rather, we rely on certain pre-reflective achievements of the body, which Joas calls a ‘body schema’ or ‘body image’, that are developed in childhood and enable us to act in particular ways in the first place. It follows that it cannot be assumed that people are capable of doing whatever they like with their bodies or that control over it is continuously exercised; there are situations where control is lost and creative solutions have to be sought (Joas,

1996:176; Joas & Knöbl, 2009:519). Moreover, this body schema is constituted intersubjectively; our relationships to our bodies are profoundly shaped by the social relations and structures in which we are acculturated (Joas & Knöbl, 2009:521).

Though this is beyond the remit of the present study, I suggest that there are a number of ways in which this conception of action can help us to understand some micro aspects of participation in genocide (see also Schneiderhan, 2013). It might help explain, for instance, variegated modes of participation or why the actions of those who partake in the ‘genocide situation’ change over time in ways that undermine the idea that certain kinds of action (i.e. killing) are constant and synonymous with ethnic identity and its intersections with region, class, gender and so on (Luft, 2015). It also sheds light on some of the affective and corporeal aspects that are missing from accounts such as Bauman’s; fear, paranoia and various other forms of psychological and bodily complexes, as well as their instrumentalisation in crisis situations, are surely all important factors in explaining the sanctioning of and participation in 32

mass violence at the micro-level (Semelin, 2007 Moses, 2011a). Additionally, it draws attention to how participatory action in genocide is constituted intersubjectively (Fujii, 2004), and holds fascinating potential for understanding the links between recurrent patterns of violence and their inscriptive, corporeal function in particular cultural contexts, such as the cutting of Achilles heels of Tutsi men in Rwanda 1994 or the narrative memorialisation of microphysics of violence in Burundian Hutu refugee camps (see Malkki, 1995:ch.2; Taylor,

1999:ch.3).

For the purposes of this study, Joas’ conceptualisation of the creativity of action is most helpful in alerting us to the situational contexts of genocide and the particular conditions under which processes led by elites, like ethnic mobilisation or the sanctioning of violence, occur.

His own oeuvre is indicative in this regard. Where Joas has turned his attention to empirical phenomena, he has been particularly concerned with how value-commitments emerge, such as those oriented towards human rights (Joas, 2000, 2013). He has also been concerned with war

(2003; Joas and Knöbl, 2013) and provides a rigorous theoretical grounding in terms of the problem of the ‘emergence’ of genocide outlined above. To quote Joas and Wolfgang Knöbl

(2009:523 see also Arnason, 1996:112) at length:

The sociological analyses of wars, their causes, development over time and consequences can do much to relativise the ideas of progress so common in sociology and particularly modernisation theory. Studying wars is also useful because they are a prime example of the impact of contingency, of the non-necessary, upon history. Wars are thus not only phases that tend to be neglected because they represent a ‘dark’ element within a process of ‘development’ that is often depicted very positively. They are also nodal points of history, because the experience of war and the consequences of wars open up unpredictable possibilities for actors. This sets in motion a vast number of new processes, which brings out the absurdity of the popular notion that history is linear. To put it in terms of action theory, the actors respond to the ‘situation’ of war by creatively generating new plans. The concept of ‘creativity’, it should be underlined, entails no normative evaluation. The creative projects that have arisen during and after wars have by no stretch of the imagination all been morally ‘good’, as is clearly evident in the common references to the birth of fascism from the spirit of the First World War. I suggest that this implicates genocide as a practice related to modernity. As I elaborate in chapter three, modernity ought to be seen not as solely reducible to institutional 33

arrangements and technological features, nor as a definable time period, but as articulating an imaginary open to a future malleable by the actions of human beings according to the expectations of the present. Modernity is subsequently open to multiple interpretations that entail creativity, but these interpretations are always situated in broader contexts of constraint.

Genocide can be a creative product of contingent situations of political and economic crises and invasion or war, but so too can the emergence of new values such as genocide prevention, which must also be construed as an immanent phenomenon of modernity.

This constitutes, I argue, a welcome supplement to the ‘modernity of genocide’ thesis in genocide studies that tends to reduce modernity to Western institutional or technological indicators, or (mis)interprets Bauman’s (1989) analysis to assume that these themselves are determining factors in genocide (e.g. Straus, 2001; Jones, 2011:424-427). This often leads to distinctions between ‘modern’ and ‘non-modern’ genocides made on account of distinctions in the mode of killing. Rwanda (up-front; physical; carried out with machetes; spontaneous; passionate; value-rational) is a synecdoche for ‘pre-modern’ genocide and the Holocaust

(distanced; detached; banal; calculated; scientific; instrumental-rational) represents the

‘modern’ variant.

In recognising the ‘creative’ aspects of modern genocide, it is clear that it is not contained within the cool, mechanistic mode of killing of ‘bare life’ in concentration camps

(Bauman, 1989; Agamben 1998). This has been recognised by Dan Stone (2004) who, drawing on Emile Durkheim (2001 [1912]), argues that modern genocides have often been manifestations of a transgressive and creative reinvigoration of communal solidarity; episodes of collective effervescence emerging as responses to the anomic conditions created by societal crises (see also, Joas et al, 2006:21; De Swaan, 2015:114). They could additionally be seen as resulting from ‘a particular interpretation of modernity [by state elites] that privileges… the mastery of political crises over the expression of collective autonomy’ (Wagner, 2015:8). In 34

short, there is no singular ‘genocidal type’, and modern genocide has multiple specific expressions. It is now necessary to address the similarly complex definitional problem of colonialism.

COLONIALISM

DEFINING THE SPECIFICITY OF COLONIALISM Just as we need a definition of ‘modern genocide’ that mitigates the shortcomings of an ahistorical and Eurocentric definition, we also need to define ‘modern colonialism’. Firstly, it is necessary to distinguish it in a semantic field of related terms. The first of these is empire, an organising term under which colonialism itself needs to be situated. In the nineteenth and twentieth centuries, empire came to refer to large-scale political organisations based in overseas territories possessed by a single state and forged by conquest. The associated term imperialism emerged at this point as a pejorative term which suggested illegitimacy and hubris (Steinmetz,

2014:79). A specifically economic definition was coined in the twentieth century; in the teleological scheme of Marxist historical materialism, following Lenin, imperialism came to refer to an advanced stage in the development of modern capitalism (Fieldhouse, 1983:2).

Empire and imperialism are broader concepts than colonialism, even if in popular usage they are often conflated. When speaking of this broad rubric under which empire and colonialism are collected, I shall use the term ‘colonial-imperialism’. Furthermore, a colony ought to be distinguished from the terms colonisation and colonialism. For Jürgen

Osterhammel (2005:4), colonisation is best defined as a process of territorial acquisition. D. K.

Fieldhouse (1983:4) suggests that the definition of colonisation entails the movement of large numbers of settlers to a new territory and thus identifies it with European settler societies such as those of the Americas, Southern Africa and Australasia. Colony refers to a particular type of socio-political institution; in the nineteenth and twentieth centuries, those under exploration in 35

the present study, this was often a colonial state or administration. Colonialism refers to a system of domination. As Osterhammel puts it:

Colonialism is a relationship of domination between an indigenous (or forcibly imported) majority and a minority of foreign invaders. The fundamental decisions affecting the lives of the colonised people are made and implemented by the colonial rulers in the pursuit of interests that are often defined in a distant metropolis. Rejecting cultural compromises with the colonised population, the colonisers are convinced of their own superiority and of their ordained mandate to rule (Osterhammel, 2005:17). Modern colonialism, then, might best be conceived in a material sense as a system in which political sovereignty in a given territory (a colony) is seized by a foreign political power with superior military and technological power (a coloniser). Economically, it entails the appropriation of resources from that territory (manpower, land, raw materials etc.) and involves the integration of the colony into transnational economic structures, including in some instances through the establishment of settlements and population transfers (colonisation). This is accompanied by discourses which construe the indigenous population to be culturally and/or biologically inferior, a fact purportedly given by both history and nature and encapsulated in the organising concept of race (Balandier, 1970:34; Mann, 2005; Steinmetz, 2008:591). This definition is a useful starting point. When trying to understand the specificity of the case under investigation in terms of a historical sociology of colonialism, however, it needs to be further elaborated upon on.

POSTCOLONIAL SOCIOLOGY AND EUROPEAN COLONIALISM In order to define this specificity, I want to discuss briefly ‘European’ colonialism as conceived in postcolonial sociology (Magubane, 2005; Costa, 2007; Go, 2013, 2013a;

Bhambra, 2007, 2014; Rodríguez, Boatcă and Costa, 2010; Ascione and Chambers, 2016). The main critical thrust of postcolonial sociology is oriented towards a consideration of the entanglements of sociology, and the knowledge subsumed under its rubric, in the historical production and maintenance of a starkly unequal world. It has tended to be directed towards generating more reflexivity about the Eurocentricity of sociology, vis-à-vis a consideration of 36

the colonial relationships that were present during the discipline’s inception and were, it is argued, constitutive of foundational events in the genesis of modern societies. Where the historical development of the West has been conceived endogenously, even in terms of being one particular variety of modernity among others, the importance of colonial relationships,

Atlantic slavery and other forms of violence, dependency, appropriation and dispossession have been unduly disavowed. This reflexivity ought to amount to a radical ‘reconstruction’ of the foundations of sociology and the master-concept of modernity, which would ultimately enable sociological work to incorporate into its remit events, processes, thinkers and ideas outside of or marginal to the West.

I do not wish in any way to underplay the influence of postcolonial theory in deconstructing the chauvinism and racism that underpinned colonial-imperialism and survived formal independence into the present, including in sociologists’ ‘entanglements’ with empire

(Steinmetz, 2013). As should be clear already, I am influenced by this position inasmuch as it seeks to transcend the problem of Eurocentrism and encourages the centralisation of colonial- imperialism in the historical formation of European modernity. However, here, I would like to point to how it falls short in providing tools and concepts for engaging sociologically with what

Georges Balandier (1966, 1970) called ‘the colonial situation’, the complex of relations between the colonising and colonised societies, and its emergence in specific time-space contexts. This paradox is manifested in, firstly, a tendency to collapse the specificities of different forms of colonial-imperial rule into a generalised and ahistorical ‘European’ or

‘Western’ colonialism18, and secondly a tendency to reduce postcolonial phenomena, in this case postcolonial violence, to the impact of the encounter with Europe or the West.

18 The notion that colonial-imperialism is part of the ‘underside’ (Dussel, 1996) or ‘darker side’ (Mignolo, 2011) of European modernity can paradoxically be presented in a Eurocentric way. Europe or the ‘West’ is generally the only colonial-imperial power considered in these analyses (for this critique 37

In the attempt to understand empirically the development of the Great Lakes region, one quickly becomes confronted with the difficulty of positing a singular ‘European’ or

‘Western’ colonial-imperial project. While colonial administrators of course made recourse to

European civilisational and racial superiority, European colonial powers were, and saw themselves as, distinct states. Colonial-imperialism in the nineteenth century, when Burundi and Rwanda were first ‘encountered’, occurred against the background of a realist state of international relations, regulated by international treaties. These colonial empire-states were in competition with each other and Europe would soon be engulfed by conflicts in its metropolitan heartland (Wagner, 2009:256; McLennan, 2010:123).

This conflict of Western nations very much implicated its hinterland, in terms of providing both proxy battlegrounds and military reinforcements. The Great Lakes region, with

Burundi and Rwanda at the centre, experienced some of the worst effects of World War I on the continent (Newbury and Newbury, 2000; Carney, 2011:51). Borders were established by

European powers through diplomatic means, the Berlin Conference of 1885 being the most famous example. Burundi and Rwanda were used as pawns by Belgium between 1916 and the

1925 establishment of the League of Nations, who valued them inasmuch as they were used as pawns to get Portugal to relinquish the south bank of the Congo estuary (Jewsiewicki,

1986:474). Borders were also reinforced through violent confrontation as demonstrated by the clashes between the military forces of German East Africa in Burundi and the force publique of Congo Free State in the late nineteenth century (Chrétien, 2003:214-220).

see Ellul, 1978:10-11; see also Arnason, 2003:343 on the specificities of the colonial modernities of Korea or Taiwan under the Japanese Meiji colonial empire). In spite of its deconstruction of the grand narratives and universals of European/Western colonialism, the critique of Eurocentrism can often seem to portray European/Western colonialism as the only colonialism of universal significance (Lazarus, 2004; Delanty, 2006:668). 38

This is not to deny that a transnational European identity can be posited (see Outhwaite,

2008). By the nineteenth century, when Burundi and Rwanda were ‘discovered’, Europe was conceived as a single historical entity, a shared living space and a supranational identity marker

(Delanty, 2013:6; Osterhammel, 2014:87-88). Among the various facets of this shared identity, we can include the concept of the European race, to which appeals were certainly made during the colonial encounter in Burundi and Rwanda and beyond, as an entity biologically and culturally superior riding on the wave of history, developing in counterpoint to a ‘constitutive outside’ against which it developed (Laclau, 1990; Dussel, 1993; Hall, 1992; Mazower, 2006).

However, colonial discourses of a supra-national European racial identity do not neatly translate into a supra-national European political identity that itself colonised. Furthermore, it is not the case that recourse to a shared European identity and heritage always meant a legitimation of racial superiority and colonial-imperial aggression. Appeals to a shared

European identity were also made with the intention of forging a space of ethical solidarity that would put an end to extra-continental racial and nationalistic aggression and hostility in the colonial territories, as well as on the home continent (see Harrington, 2016). If it is the case that the First and Second World Wars were the return of the ‘boomerang’ (Sartre, 1965:20;

Foucault, 2003:103) first cast into the slave plantations and colonial territories, this return in a large sense destroyed Europe as a unitary entity. While a conception of European structural unity might be posited out of a constitutive disunity and conflict (see Elias, 1978:ch.3; Simmel

2009 [1908]:ch.4), a ‘unity in diversity’ (Delanty, 2013:10-17), a singular, supra-ordinate and agential Imperial Europe is often reproduced in postcolonial sociology19.

19 The notion of ‘Eurafrica’ discussed by Peo Hansen and Stefan Jonsson (2014) is interesting in this regard. This refers to plans to sustain European economic dominance over Africa in the formation of the European Economic Community (EEC) in 1957. The idea of European economic integration and the material connections sustained following formal independence of the colonies of European powers is of great significance, but it does not paint a full picture of European-African relations beyond the 39

There was also large variation among modes of colonial organisation. For instance, as

George Steinmetz (2008) has shown, German Southwest Africa, German Samoa and German

Qingdao were organised according to divergent logics and structures, which ultimately had a significant impact on the experiences of the colonised societies – genocide in the case of

German Southwest Africa. In Burundi and Rwanda, for instance, the relatively weak German colonial administration, while at first maintained by the Belgians, was significantly developed by the Belgian colonial state who inaugurated the administrative union Ruanda-Urundi.

Similarly, Belgium’s colonies in central Africa were significantly less encompassing and ambitious than the British or French empires. Osterhammel defines Belgian Ruanda-Urundi, and Belgian Congo to which it was attached in 1925, as ‘colonial possessions without a colonial empire’ (Osterhammel, 2005:18), on the basis that the number of Belgian colonial ‘possessions’ does not surpass the level which we would call an ‘empire’20. Belgian colonialism, furthermore, had a specifically paternalistic character, exemplified in the attitudes of Catholic missionaries

(‘the White Fathers’) and administrators like Pierre Ryckmans (Gahama, 2001:61). It is also prudent to bear in mind that Belgium itself only emerged as a state in 1830, strongly influenced by the machinations of more powerful European states. It was King Leopold II’s concern with prestige, to ‘catch up’ and compete with other western European states, which set in motion the creation and destructive annexing of the Congo Free State (Hochschild, 1999).

economic. Though economic domination over Africa might have been supported, in the Marxist sense, by ideological legitimation of a triumphant and superior Europe, this does not mean that such postulations of intercontinental relations and supranational identity are the only ones. To reduce the idea of Europe simply to flows of capital, goods and services, themselves sustained by appropriative relationships with former colonies, is to forget competing claims on European self-understanding. It is also worth bearing in mind the close relationship between nationalistic exceptionalism and nostalgia for empire in various European countries, especially Great Britain. A significant force behind the ‘Brexit’ campaign to leave the European Union, in some sense a rejection of a shared European identity, was what Paul Gilroy (2005) identified as ‘post-colonial melancholia’, referring to feelings of loss and yearnings to ‘take back control’. 20 Belgium, with its internal divisions and neutrality secured by international guarantees, not to mention the financial costs incurred upon inheriting the Congo Free State after the international humanitarian campaign against King Leopold II, was a reluctant colonial power (Jewsiewicki, 1986:462; Ewans, 2003:167). 40

This again raises questions about the unity of Europe as a colonial power. Europe itself has internal peripheries shaped by the legacies of the Ottoman, Napoleonic, Habsburg and

Russian empires (Outhwaite, 2008:ch. 2; Delanty, 2013:ch. 10). As Arnason (2003:349-350) puts it:

The critique of Eurocentrism should be accompanied by a pluralisation of the idea of Europe – in the sense of distinctions between aspects of the European experience, each with its specific historical dynamics; regional patterns and civilisational constellations within the European world; and successive overall configurations from the early modern Atlantic world and beyond. It should also be noted that differences among Europeans have in certain historical periods been considered as racial differences (Law, 2013:7).

Ultimately, this necessitates being attentive to varieties of colonial-imperialism and its specific expressions, in a similar sense to notions of the multiplicity of modernity. This would also involve conceiving of colonialism as processual and gradational, making it necessary to account for changes in the underlying structures of colonial systems over time, such as the shift from direct to indirect rule during the nineteenth century (Mamdani, 2012), or the emergence of what Osterhammel (2005:61) calls the ‘late colonial intervention state’ after the Second

World War 21 . To again evoke Elias’s (1978) language, ‘colonialism’ can often give the impression of stasis and universality; thinking in terms of processes of colonialisation perhaps better equips us in understanding changes in colonial practices and colonised societies in history. Time and transformation become significant categories here. Burundi and Rwanda were not ‘discovered’ in 1492, when Christopher Columbus set sail for the New World. Indeed, owing to various factors including the natural bulwarks of mountains and swamps and the lateness of the incursion into central sub-Saharan Africa following the turn from colonial slavery to the exploitation of African labour (Steinmetz, 2014:83), European explorers only set

21 Osterhammel points out how, particularly in French and Belgian colonies, improvements in health and educational care were implemented in African colonies following 1945: ‘Belgian politics went furthest by introducing accident and health insurance, maternity benefits, family subsidies, and retirement pensions for their African labour elite’ (Osterhammel, 2005:61). 41

foot in Burundi and Rwanda towards the end of the nineteenth century, a time when the racial logics undergirding colonial governance were at their most developed.

Moreover, there is the related problem of the historical variation within the practice or style of colonialism itself. The New World settler colonialism of the Americas or Australia

(Veracini, 2010) should be distinguished from the sorts of indirect rule seen in Burundi and

Rwanda, as should its ramifications for relations with native populations and colonial violence.

Where settlers may have abhorred and sought to destroy difference on account of obstruction of land and resources, administrators in indirect rule colonies like Ruanda-Urundi actively produced, classified, managed and preserved difference. In short, though we can certainly accept that the definition of colonialism derives from the general by analytical necessity, this general definition has limited explanatory or interpretive value when it comes to understanding the workings and effects of colonisation, colonialism and colonies in specific time-space contexts22.

Such an understanding of colonial states also points to how they were more dynamic and unstable than allowed for by reified notions of European colonial-imperialism. Beneath unitary discourses of civilisational and racial superiority often lay fundamental material disagreements about how colonial states should be run in practice. As John Comaroff (2002; see also Steinmetz, 2014:83) has shown, the colonial state (made up of an assemblage of

22 Of course, the counter-argument would be that while colonial masters may have changed (and in Burundi and Rwanda, we see a transference of Ruanda-Urundi from the possession of Germany to Belgium), the colonial system, ‘the unchanging complex of rule, exploitation, and cultural conflict in ethnically heterogeneous political structures that had been created by influences from without’ (Osterhammel, 2005:26), remained the same. In Ruanda-Urundi, the Belgians certainly worked on the legacy of the German colonial powers, even if their assessments of their predecessors were poor, rather than totally dismantle their administrative framework to build their own in its place. This makes sense when considering the anti-colonial writings of Fanon or Césaire, which were tied to a Pan-Africanist continental political project of decolonisation from Europe. But it is not at all clear how tenable the assimilation of these ideas developed in the context of African decolonisation in the 1960s is for the empirical study of colonialism in our post-, or neo-, colonial times. 42

discursive practices, institutions, material and moral concerns, religious missionaries, government officials and business interests) was a far less coherent entity than can often be implied, in spite of its high-modernist, order-building ambitions (Bauman, 1991; Scott, 1999).

The idea of the colonial state as a behemoth, as Bula Matari in Crawford Young’s (1994) vocabulary23, downplays the extent to which colonial administrations became entangled in precolonial dynamics, and that their attempts to order the colonised population led frequently to unintended disorderly consequences (Berman, 1997). The notion of the omnipotence of colonial-imperial power also downplays the creative agency of the colonised in the transformations and endurances of the colonial experience of modernity.

GENOCIDE AND COLONIALISM

Given that the system of colonialism is predicated on a relationship of domination, that the process of colonisation through expansion and acquisition brings populations into conflict over land, resources and labour, and that the foreign dominator understands itself as superior and therefore conceives of the colonised as a lesser entity, the establishment of colonies necessarily raises the question of violence, including genocide.

Again, however, here we have to be careful not to posit too general a concept of colonial violence, or collapse specificities within it. The enormous, genocidal exploitation of the population of the Congo Free State under King Leopold II was not the same as the violence and force used in colonial rule in Ruanda-Urundi under the Belgians, or Belgian Congo for that matter, even if there are areas of continuity, endurance and denial (see Anstey, 1966, and

Mamdani, 2011, on the colonial creation of ‘ethnic homelands’, and Goddeeris, 2015, on amnesia vis-à-vis the Congo Free State atrocities in Belgium). Similarly, the genocide of the

Nama and Herero following the uprising in German Southwest Africa (today’s Namibia) were

23 Young uses the nickname of the explorer of central Africa, Henry Morton Stanley, Bula Matari, translating as ‘breaker of rocks’, as a metaphor for the African colonial state in general. 43

not replicated in German colonial rule in Ruanda-Urundi or Tanganyika, although scorched earth tactics and ‘dirty warfare’ were indeed used at times to ‘pacify’ rebellious sections of the colonised population.

Generally, this complements the definition of genocide and genocidal violence outlined above24. Here, too, genocide is a form of ‘situationally creative’ social action that is operationalised in order to effect mastery in situations of crisis, typically colonial warfare; genocidal violence was often strategically employed in the suppression of anti-colonial uprising and insurgency. This is the context in which the concentration camp as a spatial and disciplinary form emerged (Lal, 2007; Smith and Stucki, 2011). This also parallels the above arguments about the messy materiality of colonial rule. Drawing from Foucault (1977:ch.1), we might say that genocide was not the zenith or truth of colonial power, but rather a strategy employed when disciplinary power became illegitimate for the subjects that it worked upon.

This is not to say that genocide is not more common or predictable in specific types of colonialism. As writers like Patrick Wolfe (2006), Mann (2005) and indeed Lemkin25 have shown, colonial genocide was most common in the case of settler colonialism where the central conflict is between an outside labour force and an indigenous occupier over access to land and resources (see Australia, New Zealand, Canada and the Americas). Here, in the absence of a need for an indigenous labour force, there is a ‘logic of elimination’ (Wolfe, 2006) which in some circumstances could develop into genocide. Overall, however, as Osterhammel

(2014:127) writes in his global history of the nineteenth century, ‘it was not in ‘the logic of

24 See pages 27-34. 25 Investigations of Lemkin’s intellectual development convincingly demonstrate that he thought that genocide was practised in the colonies long before it was visited upon Europe (Schaller, 2005:531; Fitzmaurice, 2008). When he died in 1959, he had very nearly finished a book on the role of genocide in global history. Colonial cases featured prominently in this projected global history of genocide, containing chapters on the genocide by the Germans against the Native Africans; the atrocities in the Belgian Congo; the Hereros; Hottentots; native Americans; the Aztecs; the Yucatan; the Incas; the Maoris of New Zealand; Tasmanians; and indigenous Australians (Moses, 2008:9) 44

colonialism’ to murder the colonial subjects. They could be and were used for labour’. Indeed, this was the biopolitical logic and rationale of the ‘civilising mission’, a (sometimes extremely violent) mechanism for making the colonised population productive (Mann, 2005:35).

Our understanding of the relationship between genocide and colonialism has been advanced significantly by work in genocide studies (Zimmerer, 2005; Moses, 2008; Kuhne,

2013). Such work has built upon the earlier examples of those who had earlier argued compellingly that the processes which had led to genocide and destruction on the European continent had developed in the laboratories of the European colonies. Aimé Césaire stated that genocide was only named as such because, before the Holocaust brought it to Europe, ‘it had only been applied to non-European peoples… before engulfing the whole edifice of Western,

Christian civilisation in its reddened waters’ (2000 [1972]:37). Frantz Fanon described Nazism as ‘a colonial system in the very heart of Europe’ (Fanon 1967: 33). W. E. B. Du Bois wrote in The World and Africa (1965 [1947]) that ‘there was no Nazi atrocity – concentration camps, wholesale maiming and murder, defilement of women, or ghastly blasphemy of children – which the Christian civilisation of Europe had not long been practising against coloured folks in all parts of the world in the name of and for the defence of a Superior Race born to rule to world’ (in Gerwarth & Malinowski, 2009:280). And in her Origins of Totalitarianism, Hannah

Arendt wrote that ‘it may be justifiable to consider the whole period [of imperialism] a preparatory stage for coming catastrophe’ (Arendt 1967:123).

Thus, the Nazis’ flagrant breaking of international treaties was only deemed criminal inasmuch as its colonial-imperialism (after its overseas colonial-empire had been broken up at

Versailles) spread over the European continent rather than in Africa, Asia or Latin America

(Mazower, 2006). It has been observed that there are many continuities, including personnel, the racial experiments, and modes and logics of containment in camps, between the Herero genocide and the Holocaust. More broadly – and more significantly – this argument has dealt 45

a blow to the sonderweg thesis; the German ‘special path’, it is argued, was only exceptional inasmuch as it took a widespread European/Western practice of violent colonial-imperial subjugation and turned it against itself (Varcoe, 1998:58; Fitzpatrick, 2008; Gerwarth &

Malinowski, 2009).

Another dynamic in the colonial relationship is relevant to our discussion here, namely that of the violence that accompanied decolonisation. Shaw (2011) highlights the

British role in the violence of the partition of India, as well as the genocidal counter-insurgency operated against the Mau Mau in Kenya (Shaw, 2013:108-109). Kuper (1982:ch.4) discussed decolonisation at length, including the example of French Algeria and the interethnic wars in

Nigeria during the 1960s. These cases are not limited to independence struggles from European powers; the Indonesian occupation of East Timor from 1975 and decolonisation in 2003 has frequently been interpreted in terms of genocide (Kiernan, 2009; Dunn, 2013).

A more controversial notion is that of ‘subaltern genocide’ (Moses, 2008; Jones and

Robins, 2009). Genocide, it is said, has in some cases been the result of the humiliations and frustrations of a subjugated population rising up against their violent oppressors.

Decolonisation, along with slave uprisings and peasant jacqueries, has been identified as a particularly fruitful field of study in this regard, and references are made to Fanon’s (1965:1) declaration that ‘decolonisation is always a violent phenomenon’ and Jean-Paul Sartre’s (1965) evocations of the ‘cleansing’, restorative nature of anti-colonial violence. Many have argued that Sartre’s preface to Fanon’s The Wretched of the Earth obscures the ambivalence of the latter’s thinking on violence (Gordon, 1996; Bernstein, 2013), and some of the work in genocide studies conflates the two in problematic ways (Jones, 2009:53-54). Additionally, there is seemingly little reflexivity regarding the normative problems inevitably entailed in attempting detached formulations of concepts like subaltern genocide within academic departments in ex-colonial centres. 46

There are nevertheless important themes in this literature. The first is that there is undoubtedly a perception among many perpetrators that they are an oppressed or colonised group and that violence is the only way to rid themselves of humiliating subjugation in many genocides, including in Burundi and Rwanda, as we will see (Mamdani, 2001). The second is that these acts are often couched in the discourse of modernity: liberation; progress; freedom from the shackles of feudalism and serfdom; and so on (Hinton, 2008:445).

Genocide prevention and anti-colonialism, I suggest, also have to be considered as expressions of modernity. As has been detailed, Bauman’s (1989) most illuminating and disturbing conclusion from his study of the Holocaust, a point often missed by his interlocutors in genocide studies (e.g. Jones, 2011:424-426), is that the Holocaust possessed a future- orientation; it was a purposive movement towards a ‘grand design’. Bauman gives no consideration, however, to how it might be said that a modern future-orientation also provides the resources for articulating disgust and outrage at the occurrence of the Holocaust and, beyond this, the institutionalisation of the Declaration of Human Rights and the 1948 Genocide

Convention.

These were not simply guilt-reactions to ‘disaster triumphant’ in Europe following the defeat of the absolutist and totalitarian movements of the Second World War, but emerged from a long history of progressive, humanitarian and anti-colonial critique (Fitzmaurice, 2008).

Thus, whilst we should be critical of the gap between discourses of prevention and human rights and their actual materialisation in the world, and mindful of how particular interests can be smuggled behind their universalistic appeals, these cannot be simply written off as masks of domination. They are achievements of modernity. How else can we describe the invocation

‘to liberate mankind from such an odious scourge… condemned by the civilised world’ contained in the preamble to the Genocide Convention (1948)? An interpretation of the future- orientation of modernity is entailed here which envisages and works towards a world free of 47

genocide. If a confluence of processes and logics of modernity made the Holocaust and colonialism possible, it is equally the case that other processes and logics of modernity also made them a synecdoche for evil (Alexander, 2009; Heller, 2010:109; Joas, 2013:69-94;

Harrington, 2016:44-45).

I will now move on to discuss a theory of modernity that stresses the multiplicity of its expressions. This multiplicity concerns not only the entangled historical trajectories to and through modernity, but also that which is inherent in the idea of modernity conceived as a ‘field of tensions’ (Arnason, 1991; Kaya, 2004). Genocide and genocide prevention, colonialism and anti-colonialism, and their entanglements, are expressive of this dynamic tension.

CHAPTER THREE – THEORISING THE MULTIPLICITY OF MODERNITY: TOWARDS ENTANGLED HISTORICAL ROUTES TO AND THROUGH MODERNITY

In the previous chapter, I defined genocide and colonialism and expounded some of their interconnections. Here, I look to weave these into a theorisation of modernity. In order to do this, I will first distinguish modernity from associated concepts of ‘modern society’ and

‘modernisation’. Drawing on theorists like S. N. Eisenstadt (1996a, 1999, 2000, 2001, 2003,

2005, 2005a), Göran Therborn (1995, 2003, 2011), Peter Wagner (2008, 2009, 2012), and J. P.

Arnason (1991, 2001, 2003) in dialogue with postcolonial and decolonial theorists (Said, 1978;

Mudimbe, 1988; Hall, 1992; Dussel, 1993; Mbembe, 2001; Bhambra, 2007, 2014; Mignolo,

2011), I will argue that modernity is best conceived as a time-orientation, open to a future conceived of as a space for projecting possibilities within the present. These possibilities are situated within a field of tensions and require active, creative and often conflictual interpretation. They are formed against the background of distinct historical experiences such as colonialism or revolution, and socio-cultural traditions, and unfold processually along specific historical trajectories or routes.

Modernity is thus a constitutively multiple phenomenon and appears in a variety of specific forms. This corrects earlier ethnocentric assumptions about the universality of Western modernity. I suggest, however, that distinct articulations of modernity do not proceed in isolation. Instead, they are entangled both with one another and in transnational relations of power. I will introduce the framework of ‘entangled historical routes to and through modernity’

(Therborn, 2003; Arnason, 2003a; Randeria, 2006) and delineate a typology of entanglements, before situating the empirical cases of Burundi and Rwanda within this framework. This provides an organisational model for the substantive analyses in chapters five to eight.

MODERNITY, MODERNISATION, MODERN SOCIETY

It is sometimes said that while sociology has historically been concerned with so- called ‘modern societies’ – those north-Atlantic industrial societies that emerged and experienced rapid change in the nineteenth century – colonial and postcolonial societies, typically associated with the ‘traditional’ and seen as comparatively static, have tended to fall to anthropology (Bhambra, 2007a:875; Touraine, 1989:5). Moreover, it is suggested that sociology as a discipline was born in these modern societies in order to respond to the ruptures of their industrial, political and scientific revolutions (Giddens, 1971:xi-xvi; Skocpol, 1984:1;

Vandenberghe, 2009 [1997]:1).

Modernity, then, is deemed both the study object of sociology and the key condition of its emergence. The widespread sociological use of the term ‘modernity’, however, is a much more recent phenomenon than this historical context suggests. Though its etymological roots reach into antiquity and the term itself was first used by Charles Baudelaire to describe the

‘fleeting’, ‘transitory’ nature of metropolitan life in fin de siècle Paris (Habermas, 1981:3-5), the term became widespread in social and political theory only from the 1980s. By this time, discussions abounded about whether modernity had been exhausted in a transition to postmodernity, or had transformed into a ‘late’ (Giddens), ‘reflexive’ (Beck) or ‘liquid’

(Bauman) phase (Wagner, 2001, 2012:11-13; Arjomand, 2014).

Confusion about the definition of modernity and a proliferation of vague, sweeping references to it have led to an intensive questioning of its suitability as an analytical concept fit for social-scientific or historical research (Latour, 1993; Kilminster, 2007:4; Moses, 2008;

Thomas, 2011). It is most often deployed as a master-concept referring to ‘an era and a set of institutions’, to borrow Wagner’s (2008:8) expression. A typical account of the spatio-temporal beginnings of modernity is presented by Anthony Giddens (1990:1): ‘modernity refers to modes of social life or organisation which emerged in Europe from about the seventeenth 50

century onwards and which subsequently became more or less worldwide in their influence’

(see also Eisenstadt, 1973:259; Bauman 1987:51). What had been internal to Europe, it is said, came to have worldwide influence through processes of globalisation and diffusion.

This reiterates a point initially made in earlier references to ‘modern (industrial) society’, generally in the singular, or to ‘modernisation’ in the sense deployed in modernisation theory. Modernisation theory was a post-World War II phenomenon that enjoyed particular prominence in the USA in the 1950s and 1960s (Harrison, 1988). It has roots in the ideas of

Talcott Parsons but deviated from them (and, for some, simplified them) in important ways

(Arnason, 1996:102; Joas and Knöbl, 2009:308-339). Though a complex paradigm rather than a coherent theory, the general underlying theme here was that North-Atlantic societies, led by the USA, had reached the stage of ‘advanced industrial society’ or ‘modern society’ due to the realisation of a number of core processes defined as ‘evolutionary universals’. These societies, conceived in the framework of the liberal-democratic, industrialised nation-state, possessed high levels of functional differentiation, universal value-generalisation and achievement- related roles. Societies elsewhere, in the ‘Third World’, had yet to undergo modernisation and development, and lagged behind those that had reached the achieved state of modern society; they remained ‘traditional’ societies. Over time, this model of modern society would be disseminated throughout the world via processes of exchange, diffusion and adaptation.

Deviations or ruptures from this linear process were not anticipated, and all societies could essentially be plotted on a historical continuum from traditional to modern (Wagner, 2001;

Knöbl, 2002; Joas and Knöbl, 2009:308-338; Spohn, 2011; Bhambra, 2014:17-60).

Modernisation theory had a genuinely interdisciplinary appeal, influencing historians, political scientists, economists and psychologists, in addition to sociologists. It also had a practical orientation, in the sense that its insights were seen as capable of steering developmental processes in non-Western societies (Joas and Knöbl, 2009:310). This ought to 51

be situated in relation to the Cold War 26 context as well as the emerging independence movements in South America, Africa and Asia (Wagner, 2001:37). As such, one can observe modernisation theory’s influence in scholarship on Burundi and Rwanda throughout the 1950s and 1960s, principally in the efforts of anthropologists to delineate the nature of ‘traditional’ societies and their ‘receptivity to change’ (Maquet, 1954, 1961; Albert, 1960; Louis, 1963; for discussion see Chrétien, 2003:26).

From around the 1970s, these theories of modernisation faced significant challenges

(Wagner, 2008:7). Alongside more longstanding theories of economic dependency and world- systems analysis (Rodney, 1972; Wallerstein, 2004) emerged the philosophy of postmodernity which aimed its critique at the universalising ‘metanarratives’ implicit in the linearity of modernisation (Lyotard, 1984 [1979]). These ideas influenced the development of postcolonial critique, which sought at once to identify the omission of experiences of colonial-imperialism in theories of modernisation, and to demonstrate how colonial-imperial logic was sustained and reproduced in the discourses of social-scientific and humanities disciplines (Said, 1978;

Mudimbe, 1988; Hall, 1992; Bhambra, 2014). Furthermore, developments in anthropology identified and problematised the ‘denial of coevalness’ (Fabian, 1983) implicit in notions of modernisation, whereby large parts of the world deemed to be ‘catching up’ with purportedly superior or advanced industrial societies were said to exist at an earlier stage of historical development. Modernisation theory was also undermined by events like the 1968 protests and its alleged complicity in neo-colonialism, the economic crises of the early 1970s and the subsequent transformative processes of post-industrialisation and globalisation (Wagner,

2001:33; Joas and Knöbl, 2009:312-313).

26 This also situates Francis Fukuyama’s (1989) triumphalist declarations of the ‘end of history’ following the disintegration of the Soviet Union, which revived certain tenets of modernisation theory (Tiryakian, 1991). 52

Connected to these developments is a critique that profoundly disturbs the idea that modernisation involves a straightforwardly gradual progression towards more peaceable forms of cohabitation within and between societies, even if that progression involves setbacks. This critique draws on themes presented most forcefully in the Frankfurt School tradition of critical theory (Marcuse, 1991 [1964]; Adorno and Horkheimer, 1997 [1944]). Here, it is not only the case that modern societies are not necessarily peaceful societies; they in fact contain within themselves a potential for ‘self-cancellation’ (Wagner, 2009:252) which undermines the core premises of modernisation (Therborn, 2000:62; Kaldor, 2007; Malešević, 2010; Coker, 2015).

Perhaps the most profound critiques of the ‘dream of a modernity without violence’

(Joas, 2003:29) have been presented in studies of genocide. Here, some of the most fundamental institutional and processual touchstones of modern societies have been implicated in genocidal practices, implications that deal a severe blow to the notion that they represent normative progress and as such a model to be followed by non-Western societies. Variously, the modern institutions and logics of democracy (Mann, 2004), bureaucracy, instrumental rationality and scientism (Bauman, 1989; Traverso, 2003), utopianism (Bauman, 1989), the rise of the world-system of nation states (Levene, 2005, 2005a), organic nationalism (Griffin,

2007), biopolitical logics of governing populations (Agamben, 1998; Foucault, 2003:257), and the expansion and consolidation of global capitalism (Sagall, 2013) have all been argued to possess a ‘dark side’. As a result, even if not productive of genocide in a causal sense, these have failed to prevent – and in some instances have themselves facilitated – genocide in the modern period.

MODERNITY, COLONIALISM AND EUROCENTRISM

The colonial-imperial projects of European states are also entangled in this dark side of modernity. The history of European expansion, many authors have argued, should be seen at 53

least as a product of modernity, if not central to its organisation (Dussel, 1993; Mbembe, 2001;

Bhambra, 2007, 2014; Mignolo, 2011). Thus, for Enrique Dussel (1993) and Tzvetan Todorov

(1992 [1982]), modernity begins not in the seventeenth century as Giddens suggests, but with the voyage of Christopher Columbus to the Americas in 149227. Thus, ‘modernity is, in fact, a

European phenomenon, but one constituted in a dialectical relation with a non-European alterity that is its ultimate content’ (Dussel, 1993:65). With the exploration of this alterity, inaugurated in the Age of Discovery, ‘Europe could constitute itself as a unified ego exploring, conquering, colonising an alterity that gave back its image of itself’ (Dussel, 1993:66). So much for a separation between the ‘traditional’ non-Western world and the ‘modern’ societies at the centre; on the contrary, the concept of modernity is only comprehensible in relation or connection to its traditional Other (Said, 1978; Mudimbe, 1988; Hall, 1992:187; Dirlik,

2003:279; Bhambra, 2014).

The social-scientific disciplines and their historical entanglement with colonial- imperialism are subsequently argued to have produced a Euro/Western-centric world picture which naturalises inequality and exploitative relationships between the global North and

South28. In a material sense, the processes of modernisation and the achieved state of modern society in the West came into being in large part because of exploitative and appropriative economic relationships involving the violent subjugation of other parts and peoples of the world. As Fanon put it: ‘The well-being and the progress of Europe have been built up with the

27 The pitfalls of periodisation here lead ironically to a kind of Americentrism; a decade prior to Columbus’s ‘discovery’ of the Americas, the Portuguese explorer Diogo Cão arrived at the mouth of the Congo River on West Africa and encountered the Kongo kingdom (Balandier, 1968:2). 28 It must be acknowledged that, while this was undoubtedly a feature of some or even most social- theoretical treatments of the non-Western world, it is not the case that it suffuses classical sociological thought as such. It does not follow that the discursive formation of the West in relation to alterity always had to be framed in hierarchical and normative terms. The self-consciousness of the west vis-à-vis the East, Africa or the Americas, as espoused in the works of sociologists and their forebears, could just as much be about establishing the specificity of the West, which is different from asserting its unique supremacy (see Kurasawa, 2000, 2004; Harrington, 2016). 54

sweat and the dead bodies of Negroes, Arabs, Indians and the yellow races’. Europe, he said, was ‘literally the creation of the third world’29 (1965:76, 80-81).

Such points are central to the general critique of Eurocentrism, another essentially contested concept. As I employ it presently, it has two components, each of which extends beyond locating Eurocentrism in the investigation of European subject matter or any claim emanating from European social thinkers and researchers.

Firstly, Eurocentrism refers to ‘a typical recurring pattern of defective Western self- reflection, whereby a claim is made about a state of affairs holding universally on the global stage that in fact only holds for limited European circumstances’ (Harrington, 2016:38). As we have seen, modernisation theorists held that the interconnected societal developments which gave rise to Western modern societies were evolutionary universals and that each society would eventually converge on such a model.

Secondly, ‘Eurocentrism is the belief, implicit or otherwise, in the world historical significance of events believed to have developed endogenously within the cultural- geographical sphere of Europe’ (Bhambra, 2007:5). Again, a key tenet of modernisation theory was the idea that this convergence towards a single societal model would ultimately be the result of endogenous processes (Harrison, 1988:9). World regions and civilisations have not and do not develop within hermetically sealed borders, but do so in the context of highly asymmetrical relationships with others, including colonial relationships (Randeria, 2006:215).

29 Bhambra (2011:680) uses the example of the cotton industries (pivotal to the industrialisation of cities like Leeds, where this thesis was written) to make this point: ‘cotton first came to Britain from India, as did the knowledge of how to design, weave and dye it. It was grown in the southern states of the US by Africans who had been transported there as a consequence of the slave trade. It was then shipped to the mills in Manchester and Lancaster and the fabric produced was in turn distributed to markets further afield – many of them were opened up to British goods through force and the subjugation of local populations and their activities. By addressing the Industrial Revolution from this wider perspective, we see that it is not something that emerged endogenously within Britain, or even Europe, but rather that it had global, and colonial, conditions of emergence’. 55

MODERNITY AS A TIME-ORIENTATION

Something vital is missed if modernity is treated solely as an analytical category denoting an achieved state or set of institutions. In addition to the ambiguities around the spatial and temporal dimensions of modernity outlined above, tensions also abound in terms of the narration of modernity from a normative point of view (Bernstein, 1986; Ricoeur, 1980;

Wagner, 1994:3-19). Beyond designating an era or a group of institutions, modernity entails the creative pursuit of instituting a meaningful cosmos out of the raw material of chaos bequeathed by the ‘disenchantment of the world’ (Weber, 2008 [1919]:51; see also Eisenstadt,

2005:32). Thus, modernity is almost always associated with metanarratives of emancipation, autonomy, freedom or progress, or their obverse in conformism, mastery, oppression or regression.

I suggest that postcolonial sociology and decolonial theory do not adequately confront this normative aspect of modernity. Rather, they present an ‘overcritique’30 (Kilminster, 2013) whereby modernity appears only as domination and violence (Seth, 2016). In decolonial theory, modernity and colonialism are so inextricably connected as to warrant the neologism modernity/coloniality (Mignolo, 2001; Escobar, 2007). Modernity as ‘modern society’ is defined as an era and a set of institutions as in earlier modernisation theories, except with the caveat that these institutions are largely regressive rather than progressive, and that the era begins not with the ‘triple revolutions’ within Europe but with the ‘discovery’ of the non-

European world.

This overcritique can be situated in contemporary debates in social theory regarding how modernity ought to be theorised, if at all. To paraphrase Ibrahim Kaya (2004:15), the central diagnostic issue regarding modernity seems to be the argument over whether modernity

30 Overcritique, to quote Kilminster, refers to ‘the relentless, gloomy and often melodramatic depiction of modern society as wholly noxious and oppressive’ (Kilminster, 2013). 56

is ‘exhausted’, as the postmodernists claimed, or whether it is instead an ‘unfinished project’

(Habermas, 1996). The history of colonial violence and oblivion sketched briefly above might lead us to adopt the former view that modernity – along with correlatives like progress, development, civilisation, etc. – is merely a mask used to cover up violent domination

(Bernstein, 1986; Habermas, 1987a:4), nowhere more hypocritically applied than in the colonial-imperial ventures of European nations. Modernity ‘radiates disaster triumphant’, to use the terms of Adorno and Horkheimer (1997 [2004]:3), and is therefore a dead project which ought to be abandoned.

This totalising perspective has been subject to criticism. Several sociologists have convincingly argued that the critique of positivistic and ethnocentric trends in social-scientific thought skirts over too many grey areas and important mediating positions31 (Joas, 1998;

Kurusawa, 2004; Harrington, 2016). These positions anticipate some later thinking on the multiplicity of modernity, and refute totalising, linear conceptions in both affirmative and critical modes. Such conceptions of modernity downplay its open-ended, tensional and antinomical constitution. As Arnason (2003b:35; see also Eisenstadt, 1999a:286) has argued, this is ‘a central but undeveloped theme of classical sociology’. For instance, Durkheim saw in processes of individualisation both the possibility for novel kinds of morality and solidarity – indeed, he proposed the sacralisation of the individual as the ‘religion of modernity’ – and a damaging erosion of the limits placed by society on human desire, giving rise to egoism and anomie (Durkheim, 2006 [1897]:269-278; Joas, 2013:49-64). Weber wrote of the tensions between religious and cultural orientations and institutional orders (Weber, 1993 [1922]:207-

222). Such themes are present in later significant theoretical work. For Jürgen Habermas (1984,

31 See for instance Hans Joas’ (1998) highlighting of American pragmatism in his essay on the German reception of Zygmunt Bauman’s study of the Holocaust; Austin Harrington (2016) on cosmopolitan and hermeneutical thinkers in Weimar Germany who sought to self-relativise Europe from within; and Fuyuki Kurasawa (2004) on the ‘ethnological imagination’ that he argues to be present in Western sociological thought since its institutionalisation, with precedents in Montaigne and Rousseau. 57

1987), for instance, the claim that the Enlightenment resulted in the profusion of instrumental rationality and spiritual disenchantment ought to be tempered by a consideration of the communicative rationality of the public sphere that is also part of the Enlightenment legacy.

Peter Wagner (1994, 2008), following Castoriadis (1987 [1975]), similarly suggested that modernity entails a ‘double imaginary signification ‘consisting of a tension between tendencies towards radical democratic autonomy and liberty on the one hand, and discipline and rational- mastery on the other. Modernity – and colonial modernity is surely the paradigmatic example of this – inaugurates ‘a history of oppression while at the same time serving as a reference- point for resistance to oppression and struggle for justice 32 ’ (Wagner, 2015:2; see also

Alexander, 2006; Harrington, 2016:44).

This is indicative not simply of a polarised, Manichean struggle between the forces of darkness and light, but also of a ‘field of tensions’ (Arnason, 1991) which produces the dynamic of modernity and modern societies. As Glenda Ballantyne (2008:56) puts it,

‘modernity is conceived not as a particular institutional structure, but rather as an open-ended set of ‘problems’ – the search for certain knowledge, for political order, for means to conceive the continuity of the self – which are open to diverse and changing institutional responses’. It is this interpretive dimension that tends to get missed in both the overcritique of modernity and the structural-functionalist unfolding of a systemic logic characteristic of modernisation theory.

Here, there is a tendency to ascribe agential and anthropomorphic powers to modernity itself,

32 A comparison of the Congo Free State ruled by King Leopold II of Belgium and the revolutionary slave-uprising of Haitian independence is exemplary in this sense. Leopold II justified his venture, which had an enormous death toll, in the following terms: ‘To open to civilisation the only part of our globe which it has not yet penetrated, to pierce the darkness which hangs over an entire peoples, is, I dare say, a crusade worthy of this century of progress’ (in Hochschild, 1999:44). Yet as C. L. R. James (1989 [1938]:198) wrote, for the leader of the Haitian independence movement, Toussaint L’Ouverture, the Enlightenment ideas of liberty, fraternity and equality, rather than ideological masks of European domination, were more relevant and meaningful to the Haitian slaves. The modern social imaginary was not hypocritical in itself; rather it was the French colonialists’ interpretation of it which was hypocritical.

over and above the heads of human beings. Modernity appears as a collective subject which itself acts and possesses characteristics (see critique by Weber, 1978:14). The personification of modernity as a social actor in itself disavows the transformative, creative and intentional powers of action and its unintended consequences. It is this that gives rise to structures which possess importance for people who negotiate them at the same time as they regulate their conduct in a quasi-naturalised way (Taylor, 2004:164; Wagner, 2009; Vandenberghe, 2009:1).

In Castoriadis’ (1987 [1975]:108) words, modern society is both ‘instituted by’ collectivities and at the same time ‘institutes’ them: ‘it is the union and the tension of instituting society and of instituted society, of history made and of history in the making’.

At this point, it is possible to give a distinct definition of modernity which detaches it from the institutionalist reading of ‘modern society’ or processes of modernisation. I follow

Peter Wagner (2001:42) here:

‘Modern society’ denotes a social order that gains its modernity from a particular structural and institutional arrangement. Modernity is here seen as a given and identifiable social form. ‘Modernity’, in contrast, refers to a situation, a condition, which human beings give themselves and/or in which they find themselves. This situation is in need of interpretation; and such interpretation can always be contested. This definition and distinction is important for the analysis to follow. Modernity entails first and foremost a condition marked by a particular time-orientation and a particular conception of human agency. The future is conceived as a space shapeable in the image of the expectations and possibilities of the present, and in principle contestable, by interpretive and creative collective groups of human beings (Eisenstadt, 2000; Arnason, 2001; Therborn, 2003, 2010,

2011; Taylor, 2004; Heller, 2005; Delanty, 2006, 2013; Wagner, 2008, 2012). A gap opens up, in Reinhart Koselleck’s (2004 [1979]:255-276) terms, between the ‘space of experience’ and the ‘horizon of expectation’. As Charles Taylor (2004:6-7) has suggested, this brings it close to some modern definitions of utopia ‘which refer us to a way of things that may be realized in 59

some eventually possible conditions, but that meanwhile serve as a standard to steer by … it provides an imperative prescription’.

These expectations and possibilities also shape the relationship of these collectives to the past. Articulations of modernity are formed with reference to distinct historical experiences and socio-cultural traditions (Arnason, 2001:151; Wagner, 2008). The future-orientation can thus refer to the realisation of foundational events such as revolutions or national independence or to projects of Arcadian restoration or traditional revival. In this sense, as Arnason (1991:210) argues, Romanticism is not an opposed, alternative project to Enlightenment, but rather situated in relation to it in a ‘field of tension’, ‘a constitutive polarity of modern structures of consciousness that can express itself in highly different constellations’ 33 . Differing interpretations of the time-orientation of modernity, mediated via an engagement with socio- cultural traditions and historical experiences, give rise to distinct institutional orders. These are in turn marked by a tension between their indeterminate constitution as products of human creation and autonomy and their quasi-naturalistic objectification in observable, calculable and masterable ‘world pictures’ (Heidegger, 2003 [1938]). As a result, there are multiple ways of being modern.

MULTIPLE MODERNITIES

This position is strongly influenced by Eisenstadt’s concept of multiple modernities

(2000). Born in Poland before emigrating to Palestine in 1933, Eisenstadt trained under Talcott

Parsons. He worked in the traditions of modernisation theory and structural functionalism, but

33 Hans-Georg Gadamer (2004 [1975]:275) also highlights the relationship between the Enlightenment and the ‘Romantic retrieval of origins’ which gave rise to the historical sciences: ‘nineteenth century historiography is its finest fruit and sees itself precisely as the fulfilment of the Enlightenment, as the last step in the liberation of the mind from the trammels of dogma, the step to objective knowledge of the historical world, which stands on a part with the knowledge of nature achieved by modern science’.

departed from them in significant ways. His own biographical proximity to the turbulence of the twentieth century might be said to have framed a sensitivity to the possibility for barbarism in modern societies, as well as to non-Western interpretations of the ‘cultural programme’ of modernity, particularly in Israel, and the importance of exogenous influences and contingency in processes of modernisation (Eisenstadt, 1996a, 2005, 2005a; Spohn, 2011; Alexander,

2013:ch.3). By the mid-twentieth century, Eisenstadt had already marked himself apart from modernisation theorists and the division of labour which was said to separate sociological study of the modern from the anthropological concern with the traditional. As he put it (1949:121),

‘there is no theoretical distinction between sociology and social anthropology’.

Following the lead of another of his teachers, Edward Shils, Eisenstadt adopted an emphasis on the importance of culture and tradition, which Shils argued was inadequately thematised in structural-functionalist modernisation theory. Combining Durkheim’s sociological work on the sacred with Weber’s on charisma, Shils held that ‘certain ideas about the sacred exist in every society, including modern society. We can thus by no means assume that modernity entails and will continue to entail a comprehensive process of secularisation inevitably resulting in the dissolution of all that is sacred’ (Joas and Knöbl, 2009:316).

Straightforwardly, the implication of this is that there is no clear-cut distinction between tradition and modernity, as modernisation theorists of the day tended to assume (Shils, 1958,

1981). Tradition is not simply a nightmare that ‘weighs on the brains of the living’ (Marx,

1852), a stagnant and rigid horizon encompassing an unchanging cultural collectivity, but is a continuously reconstructed foundation upon which are built differentiated and dynamic creative interpretations of what Eisenstadt (2003:ch.4) calls the ‘cultural and political 61

programme’ of modernity 34 (Eisenstadt, 1969, 1972, 1973, 1973a). These interpretations themselves produce diverse institutional responses (see Eisenstadt, 1973; Ballantyne, 2008;

Tiryakian, 2011).

The notion of multiple modernities is thus characterised by an implicit hermeneutic sensibility and openness to the influence of tradition, in contrast to central tenets of modernisation theory which reflected more general tendencies of Enlightenment thinking towards ‘the prejudice against prejudice itself, which denies tradition its power’ (Gadamer,

2004 [1975]:270; see Silber, 2011). Traditions in this conception possess what the hermeneutic philosopher Gadamer (2004 [1975]:300-301) called an ‘effective history’, referring not only to their formative and determinate influence but also to their open-endedness, their inherent capacity to produce conflicts of interpretations, and their tendency to acquire new meanings and change direction in relation to present expectations and future anticipations (Arnason,

2001:148-154; Ballantyne, 2008:57; Delanty, 2008:49; Friese and Wagner, 2000; Wagner,

2008).

Eisenstadt’s focus in this regard is on the diverse conceptions of transcendence which emerged in what Karl Jaspers (1953 [1949]) called the Axial Age. This refers to a period around the 8th to the 3rd centuries BC, where in elite subsections of several distinct civilisations in

Persia, India, China and the Greco-Roman world there occurred a crystallisation of a tension between mundane and profane this-worldly order and the transcendent and sacred other- worldly order, and conceptions of how to institutionalise them (Arnason, Eisenstadt and

34 The definition of tradition employed in this thesis follows the long-standing position developed in pre-‘multiple modernities’ scholarship of Eisenstadt. Here, tradition is multifaceted and refers variously to ‘actual or symbolic past event[s], order[s], or figure[s] as the major focus of … collective identity’; to an entity that delineates ‘the scope and nature of … social and cultural order’; and the ‘ultimate legitimator of change and of the limits of innovation’ (Eisenstadt, 1969:453). Overall, for Eisenstadt, tradition constitutes ‘the reservoir of the most central social and cultural experiences prevalent in a society, as the most enduring element in the collective social and cultural construction of reality’ (Eisenstadt, 1972:3). 62

Wittrock, 2005; Delanty, 2008:48; Joas and Knöbl, 2009:328; Bellah and Joas, 2012). For

Eisenstadt, modernity does not represent the inevitable unfolding of a universal historical law and the unique superiority of the West as its bearer; rather it constitutes a specific trajectory contingent on the Judeo-Christian legacies of the Axial Age. Within this tradition, there emerged ‘a belief in the possibility of bridging the gap between the transcendental and mundane orders [and] … the belief in the capacity of conscious human agency to realise in the mundane orders some of the transcendental and utopian visions’ (Delanty, 2008:49).

This ‘cultural programme’, consisting of a novel conception of time and the transformative role of human action within it, demarcates modernity as a distinct civilisation

(Eisenstadt, 2001, 2005:31), which crystallised firstly in Europe around such events and political movements as the Great Revolutions, the wars of religion and the Enlightenment. In what Delanty (2008:49; see also Tiryakian, 2011:243) denotes as a ‘second Axial Age’, the civilisation of modernity spread gradually to the rest of the world, often impinged on non-

Western societies in the form of colonial-imperial domination and violence or in the teachings of missionaries (Eisenstadt, 1999a:288).

The open-ended and interpretive nature of the cultural programme of modernity proliferated in multiple forms. In dialogue with the Confucian tradition, there emerged a Sinic or East Asian interpretation, and in line with the Buddhist and Hindu traditions there emerged an Indic civilisational interpretation. As such, for Eisenstadt, the study of historical processes, contingent events and orientations to the world is vital for the sociological analysis of the diversity of the present global age. Such an analysis ties into the discussion of the interpretive dimension of modernity. The diversity of modernity and its multiple institutional arrangements is to a great extent attributable to the interpretive and creative action of individual agents and collectives, a theme in which we can detect the influence of another of Eisenstadt’s teachers,

Martin Buber (Buber, 1992; see also Joas and Knöbl, 2009:319-320). 63

The essence of the idea of multiple modernities is that it stands in opposition to the long-held notion that ‘the cultural programme of modernity as it developed in modern Europe and the basic institutional constellations that emerged there would ultimately take over in all modernising and modern societies; with the expansion of modernity, they would prevail throughout the world’ (Eisenstadt, 2000:1). Instead, Eisenstadt (2000:2) argues that the notion of multiple modernities allows us to understand the contemporary world as:

A story of continual constitution and reconstitution of a multiplicity of cultural programmes. These ongoing reconstructions of multiple institutional and ideological patterns are carried forward by specific social actors in close connection with social, political, and intellectual activists, and also by social movements pursuing different programmes of modernity, holding very different views on what makes societies modern. Through the engagement of these actors with broader sectors of their respective societies, unique expressions of modernity are realized. Such interpretations lead to conflict and antinomical tendencies in modern societies, centred on the contradiction ‘between the basic premise of the cultural programme of modernity and the major institutional developments of modern societies’ (Eisenstadt, 1999:65).

There is thus, as Ibrahim Kaya (2004:17) states, ‘no single, agreed idea of modernity, but there is space which provides opportunities to interpret ‘imaginary significations of modernity’ in multiple ways, and this space is culture, or language, or history, whose importance is no less than that of power or rationality’. The notion of ‘imaginary significations’ is borrowed from Cornelius Castoriadis (1987 [1975]) and refers to imaginary components and animating ideas which are a central feature of the self-institution of every society; an imaginary signification is ‘a future-oriented projection of the possibilities within the present’ (Delanty,

2013:19, see also Delanty 2008:49). Beyond this, Charles Taylor (2004:2, 183) argues that a social imaginary is not merely a set of ideas (as in ‘ideology’) but rather ‘is what enables, through making sense of, the practices of society’. For Peter Wagner (2008, 2012), following

Castoriadis, the cultural programme of modernity is centred on the tension between the 64

imaginary significations of autonomy and mastery, while Delanty (2013:19) describes a number of imaginary significations:

There is… not simply one transformative idea, as Wagner claims, such as autonomy, but many, and it is often the combination of these ideas that make possible the specific form of a particular variant of modernity (e.g. the liberal, socialist or republican variants of political modernity were products of different appropriations of the ideas of freedom, equality, autonomy, self- government, individualism, and social justice). These imaginary significations are often in tension with one another and produce confrontations between pluralistic and totalistic tendencies, particularly in the context of the tension between individuation and the constitution of collective identities (Eisenstadt, 1998).

For Eisenstadt (2005a:643-653), one of the most significant characteristics of the shift in conceptions of time and human agency was, using an expression from Claude Lefort, the

‘loss of markers of certainty’ (Eisenstadt, 2004:395), accompanied by continual attempts to reinstate them. Collective and individual identities were no longer taken as given or preordained by a transcendental authority or by perennial custom. Rather, they were the site of continual struggles and contestations borne by distinct social actors – political activists, intellectuals, social and nationalist movements – and were directed towards the constitution of new collectivities and the mobilisation of wide sections of the populations thereof.

There developed an intensive connection between the constructions of the political order and of the major ‘encompassing’ collectivities, epitomised in the model of the nation state and the mechanism of citizenship. This was complemented by a strong emphasis on cultural-political homogeneity of the population within territorial boundaries. Civilising offensives of the sorts delineated by Elias (1994 [1939]) and Foucault (2003), operative through educational institutions, major media and the military, played a central role in the construction of this homogeneity. There also arose a self-perception of society itself as

‘modern’, in distinction to (often colonial) Others (Said, 1978; Mudimbe, 1988; Dussel, 1993).

The distinctive visions of new modern collectivities entailed the promulgation of distinctive 65

collective memories. Collective identity thus has a narrative component, in which often

‘sacred’, universalistic components of the cultural programme of modernity meet particularistic, national ones that emphasise historical, territorial and cultural specificity.

Destructive potentialities became most fully manifest in the sanctification of violence, terror and wars in the French Revolution, later in the Romantic movement, and in the combination of such sanctification with the construction and institutionalisation of nation states (Eisenstadt,

2005a).

Eisenstadt was thus acutely attuned to the destructive and traumatic possibilities of modernity and related Jacobin and fundamentalist programmes (Eisenstadt, 1996a, 1999, 2005,

2005a; Alexander, 2013:ch.3). To quote at length from his programmatic essay on multiple modernities (2000:25):

Illuminating and describing the essentially modern character of new movements and collective identities, charting courses somewhere beyond the classical model of the territorial, national, or revolutionary state, does not necessarily lead us to take an optimistic view. On the contrary; the ramifications are such as to make evident the fragility and changeability of different modernities as well as the destructive forces inherent in certain of the modern programmes, most fully in the ideologisation of violence, terror, and war. These destructive forces, the ‘traumas’ of modernity that brought into question its great promises – emerged clearly after World War I, became even more visible in World War II and in the Holocaust, and were generally ignored or set aside in the discourse of modernity in the 1950s, 1960s, and 1970s. Lately, they have re-emerged in a frightening way – in the new ‘ethnic’ conflict in parts of the Balkans (especially in the former Yugoslavia), in many of the former republics of the Soviet Union, in Sri Lanka, and in a terrible way in such African countries as Rwanda and Burundi. These are not outbursts of old ‘traditional’ forces, but the result of the ongoing dialogue between modern reconstruction and seemingly ‘traditional’ forces. The multiple modernities paradigm represents a considerable advance and extension of theorisations of modernity, incorporating postcolonial critiques of the totalising and

Eurocentric tenets of earlier conceptions of modernity. There are, however, some problematic areas, some of which have been addressed by subsequent treatments by Arnason, Wagner, and

Therborn, as detailed below. 66

The first pertains straightforwardly to my empirical concerns in this thesis. In the focus on the Axial Age civilisations and non-Axial historical societies like Japan (Eisenstadt, 1996),

Sub-Saharan Africa has remained more or less ignored. In spite of openness to Africa as a

‘frontier-region’ for civilisational analysis and multiple modernities (Arnason, Blokker &

Delanty, 2007:127), such an omission would seem to tie in with the longstanding disavowal of

African societal organisation and world-orientations; the identification of Africa merely with the lack or absence of civilisation or modernity35 (Diop, 1989; Mbembe, 2001:8; Ndlovu-

Gatsheni, 2013:xi).

This omission points to the related issue of a lack of a discussion of colonial- imperialism and its relationship to capitalist expansion (Dirlik, 2003; Joas & Knöbl, 2009:329-

330; Bhambra, 2014:19-38). Colonial modernity was a violent imposition that deprived those colonised for a significant time of the opportunity for any kind of interpretive autonomy, or indeed political and civil rights. When such interpretations became possible in the mid- twentieth century, among elites educated in the colonial language and educated in colonial schools, they were curtailed by structures of economic dependency and political interventions on the part of ex-colonial states and global institutions.

A related issue here is an overly stable and continuous notion of long-standing tradition.

Peter Wagner (2014:295-296, 2015) has recently argued this point, with reference to Louis

Hartz’s (1969) ‘fragment theory’ of the settler-colonial societies of the New World. Societies such as the USA, Brazil, South Africa or Australia cannot claim a coherent civilisational trajectory and stable cultural resource in the sense implied in the case of Axial Age civilisations.

35 This may be an unfair charge at Eisenstadt specifically. He wrote on African societal modernisation at an earlier stage in his career (1954, 1965) and recognised that experiences of mass violence in post- independence Burundi and Rwanda constitute ‘traumas of modernity’ (2000; 2005; 2005a), a term which I explore in further detail in chapter eight. He also co-edited a book on political clientalism with Réne Lemarchand, referenced widely in this thesis as an Africanist with particular expertise on Burundi and Rwanda (see Eisenstadt and Lemarchand, 1981). 67

Yet here we can see the institutionalisation of modern societies, ‘instituted in processes in which the features of the European interpretive ‘fragment’, of the relation to Europe (‘the imperial relation’), and the encounter with the native populations (‘the racial encounter’) come together to form something new’ (Wagner, 2014:300). The indirect variant of colonial modernity, as experienced by Burundi and Rwanda, is different still. Here, colonial administrators ruled through customary authorities, often according to a racialised conception of their precolonial traditions. Whereas modernity in the European metropole entailed a movement towards processes of individualisation, ‘it recast and cemented ethnic, racial, and other particularistic differences in the colonies’ (Randeria, 2006:60). As Wagner (2009:256) writes, evoking numerous postcolonial commentators, ‘European history as colonial history establishes precisely the relation between Europe and other parts of the world as relations between ‘modernity’ and ‘tradition’, of rupture in temporality and the ‘denial of coevalness’’

(see also Said, 1978; Fabian, 1983; Mudimbe, 1988; Hall, 1992; Randeria, 2006:60). I return to these issues in chapters five and six.

A further problem is the focus on societal elites as agents of modernisation (Taylor,

2004:30). Not only does this disavow the role of broader sections of the population, such as the peasantry, in modernisation, but the motivations and discourses of these groups surrounding events like revolutions might differ significantly from those of elites (Joas and Knöbl,

2009:330; for a specific treatise on this theme in relation to Rwanda, see Newbury and

Newbury, 2000). Also important here is how the language of modernity might be captured by elites in order to cement and legitimate patronage privileges and particular interests. This is particularly pertinent in the case of postcolonial African states. In chapters seven and eight, I look at how traditions and foundational events such as revolution and national liberation have been discursively employed by postcolonial elites, with the aim of controlling the broader population or for projecting an external image to aid donors. In a sense which echoes Habermas’ 68

(1988 [1970] critique of Gadamer, tradition is not a value-neutral repository for foundations of modernity that can be drawn upon by society tout court, but can be deployed to further particularistic interests (Bleicher, 1980:154; Harrington, 2001:33-38).

Perhaps the most frequent criticism of multiple modernities is its detachment from normative questions (Wagner, 2010:10-11, Harrington, 2016:359). In attempting to account for a plurality of ways of being modern, the notion of multiple modernities arguably undermines the need for universal, normative features of modernity which ought to be shared by all, regardless of their cultural variations, such as an aversion to dehumanising and oppressive legal and cultural practices, a commitment to human rights and individual autonomy, to democratic institutions, and so on. From a Marxist perspective, in its respect and tolerance for cultural difference, multiple modernities might be said to be a corollary of a sophisticated culturalist justification for inequalities attributable to global capitalism (Badiou, 2002; Dirlik, 2003). For critics like Bhambra (2007, 2014), this acknowledgement of forms of cultural difference proceeds without impacting on the assumption of the European origins of modernity itself, thus smuggling in a form of Eurocentrism even whilst claiming to disavow it. For others, multiple modernities affords too much emphasis to endogeneity, where world regions and even nation states are essentially conceived as ‘cultural containers that are coherent and bounded and reproduce themselves over time’ and form their own distinct modernities separate from those of others (Wagner, 2009:254; see also Dirlik, 2003; Joas and Knöbl, 2009:329; Bhambra, 2007,

2014; Delanty, 2008:55).

ENTANGLED HISTORICAL ROUTES TO AND THROUGH MODERNITY

There is a difference, albeit a subtle one, between conceiving of modernity in plural terms (i.e. as ‘modernities’) and stressing the multiplicity of modernity. By doing the latter, it is possible to mitigate some of the problems outlined above. To speak of the multiplicity of 69

modernity entails a global unity in which are contained specific processual trajectories. The most fruitful conceptualisation of this, which I follow in the analysis to come, is the idea of entangled histories. This was introduced by the historian Shalini Randeria (as

Verflechtungsgeschichte – with Conrad, 2002; see also histoire croisée in Werner and

Zimmerman, 2006), and elaborated in sociological terms by Göran Therborn (2003) and, to a lesser degree, J. P. Arnason (2003a). Here, there are a number of historical routes to and through modernity, which do not proceed in isolation, but are entangled with one another. Like that of multiple modernities, the conceptual framework of entanglement, particularly as utilised by Therborn, is distinct from the ‘rival idealisms of programmatic pro-modernity… and of denunciatory post-modernist anti-modernity’. In this sense, it is amenable to empirical analysis, and constitutes ‘a searchlight on actual trajectories to and through modernity, on actual forms of discourse, movements, conflicts, institutions, rather than on ideals, blueprints, and stereotypes’ (Therborn, 2003:294).

This historical framework, I suggest, can be combined fruitfully with the more cultural- hermeneutic orientation in multiple modernities, and the emphasis on interpretation. Peter

Wagner’s work offers a good perspective for historicising the hermeneutic and cultural dimensions of modernity already outlined. For Wagner (2008, 2009), such interpretations are manifestations of responses to a series of questions or problematiques, born of the commitment to autonomy; to give oneself one’s own law. He sees these as the distinctive feature, or imaginary signification, of modernity. These include how members of a given society should live together (the political problematique); how they should satisfy human needs (the economic problematique); and finally how they produce valid knowledge (the epistemic problematique).

Varieties of modernity are attributable to the difference in the interpretations of these problematiques and their institutionalisation. These interpretations are framed by and emerge from particular experiences and significant historical moments, which constitute the horizon 70

upon which specific answers to those questions are elaborated. Thus, ‘the interpretation given collectively to the experiences of those significant moments is that which gives shape to a specific variety of modernity’ (Wagner, 2008:4).

It follows that specific routes to and through modernity are not only situated in the hermeneutic horizon of socio-cultural traditions; they are also narrated with regard to foundational historical events and experiences (Wagner, 2008, 2009; Bauman, 2008:92; Nenon,

2016:249). Variations of modernity are in part attributable to the collective interpretive horizons bequeathed by specific historical experiences (Wagner, 2008, 2009; see also

Eisenstadt, 1999a:290). Charles Taylor (2004:175-178) has also noted how the idea that modernity involved a transformation in time also changed the way that history was understood and modes of narration were employed. The collective identity of ‘peoples’ such as nations and ethnic groups was grounded on new ways of narrating their stories in secular time, without recourse to notions of a self-realising order transcendent above human action. The idea of

‘progress’ in the Western imaginary can be identified here. Revolution is another key nodal point, where a people are narrated to have separated themselves decisively from age-old forms of structural oppression. As Taylor (2004:176) states, ‘the idea of revolution can easily turn into a powerful myth, that of a past nodal point whose infinite possibilities have been frustrated, betrayed, by treachery or pusillanimity’. Furthermore these modes of narration can in turn be:

interwoven with apocalyptic and messianic modes drawn from religious understandings of Heilsgeschichte (history of salvation): for instance, the idea that the maturing order must confront violent opposition, the more violent the closer it is to ultimate victory. Revolution will be attended by a titanic struggle, a secularized Armageddon. The devastating effects of this in twentieth-century history have been all too evident (Taylor, 2004:177-178). Burundi and Rwanda are in important senses representative of this aspect of twentieth-century history, as will be demonstrated in chapters seven and eight.

These interpretations, however, are always situated in a broader context of international relations and global transformations, and under external material constraints. As I will show, 71

the interpretations of the imaginary significations of modernity in Burundi and Rwanda have historically been severely constrained by their colonial experiences and their relationships of dependency to ex-colonial powers. Any future-orientation is under severe pressure to adapt to the incantations of more powerful states and of global institutions including the World Bank, the United Nations and the International Monetary Fund. A number of thinkers (Berardi, 2011;

Lazzarato, 2011) have highlighted the ‘future-cancelling’ properties of debt in Western neoliberal societies; this has arguably longer been the experience of postcolonial societies, for instance under austerity policies introduced in structural adjustment programmes.

These constraints and relationships can be addressed in conjunction with the thinking of Therborn. For Therborn (2003:295), modernity is a global phenomenon entailing finitude and connectivity, as opposed to a universal one. Modernity ought ‘to be put into a global context, but not as different modernities, which would be a descriptive cop out, but as something produced and experienced differently by different, but linked, pathways to modernity’ (Therborn, 2010:71). Modernity is better defined not as a set of institutions, ‘which would never allow you to escape from arbitrariness, about what is modern, pseudo-modern, or non-modern’, but rather in ‘non-arbitrary, etymological terms as a time orientation, a culture, an epoch, a society, a social sphere oriented to the future, as something new and makeable, disavowing the authority of the past, of tradition, questioning ancient wisdom’ (Therborn,

2010:72). These orientations can be studied empirically ‘with respect to specific practices, such as cognition, art, economics, and politics’ (Therborn, 2010:72). But, crucially, these orientations do not arise in a context of separation. The creative interpretation of the imaginary significations of modernity always operates under the conditions and constraints enforced by the entanglement with other modern societal formations, institutions and interpretations.

This enables us to think about power, including colonial power, in a way not contained within the notion of multiple modernities. For Dirlik, the route to and through modernity is a 72

relational rather than achieved state, attributable to its interpretive translation with endogenous cultural traditions: ‘modernity is not a thing but a relationship, and being part of the relationship is the ultimate marker of being modern’ (Dirlik, 2003:279). Bhambra (2007, 2014) has also made this argument. Following the terminology of the historian Sanjay Subrahmanyam, she argues for a proposed ‘reconstruction’ of the sociological organising concept of modernity that incorporates the ‘connected histories’ which have constituted it (2007:76). My preference for

Randeria’s historiographical terminology of ‘entanglements’, and Therborn’s incorporation of it, is because it avoids the tendency to reduce phenomena and historical trajectories solely to

‘relation’ or ‘connection’, which can be the case in particular with postcolonial or decolonial treatments of phenomena within the post-independence period. As David Newbury (2012:48-

49) and René Lemarchand (1994:22-26) have shown with specific reference to Burundi and

Rwanda, postcolonial mass violence and genocide in the 1990s cannot be reduced to the colonial relationship or connection begun over a century before, as has sometimes been argued.

Entanglement certainly entails a relation or connection (i.e. the convergence with other historical trajectories) but, importantly, also entails divergence along local or national lines.

TYPES OF ENTANGLEMENT I make a distinction between three forms of entanglement, each of which frames the empirical analysis in the second part of the thesis. One has to be able to distinguish what is entangled for it to be an analytically useful device:

1. Entanglements of modernity and tradition: The first are the ‘constitutive entanglements of modernity and some tradition, coming

out of the infinitely variable incompleteness of every modern rupture with the past, and

out of the plasticity of most traditions’ (Therborn, 2003:295; see also Arnason,

2003a:308). As discussed at length above, any given route to and through modernity is

interpreted within the horizon of social-cultural tradition – a ‘cultural programme’ in 73

Eisenstadt’s terms – and is narrated with reference to a background of historical

collective experience. This mitigates the tendency in social science and political

discourse to see ‘imbrications of modernity and tradition as exceptions, deviations,

Sonderwege, or at least as limitations of modernity or modernisation’ (Therborn,

2003:295).

2. Societal entanglements:

These entanglements are those which result from the ‘very different but significantly

interacting and mutually influencing socio-political roads to and through modernity’

(Therborn, 2003:295). These predominantly involve those encounters between societies

that occurred in the unfolding of the colonial relationship, in this case Rwanda’s and

Burundi’s entanglements with the successive German and Belgian colonial

administrations. To be clear, in no way is the metaphor of ‘entanglement’ intended to

obfuscate the violence and domination underpinning colonial-imperialism, a charge

that had been levelled at some multiple modernities theorists 36 . Entanglement, I

suggest, encapsulates a multifaceted idea of violence, as in Slavoj Žižek’s (2009)

theorisation of its subjective (physical) and objective (systemic and symbolic) forms37.

The entanglement of Ruanda-Urundi into the global economy was a systemically

violent one, reinforced by the physical subjective violence of corvée labour, and the

symbolic violence entailed in occidental representations of racialised, ‘traditional’

societies. Nevertheless, colonial-imperialism cannot be reduced simply to a ‘clash’

36 As Bhambra (2014:35) has noted, the vocabulary of multiple modernities euphemises European states’ colonial-imperial expansion as ‘diffusion’, ‘crystallisation’ or ‘encounter’ (Eisenstadt, 2000; Arnason, 2003:287-296) 37 Such a reading lines up alongside applications of entangled histories elsewhere. As stated in the preface of a study of the ‘entangled histories’ of the Balkan nations, whose history, tellingly, is comparably tumultuous in terms of the Great Lakes region: ‘the ‘entangled history’ (or histoire croisée) and related approaches do not aim to smooth out conflicts and harmonise the past. The contacts, movements, exchanges and transfers were more often asymmetrical and violent rather than harmonious and peaceful’ (Daskalov, 2013:xv). 74

between two incompatible social imaginaries, one of which (the European) overpowered the other (African) and therefore led to the material domination of the latter by the former. Colonial modernity may well have arrived in large parts of the world ‘out of the barrel of a gun’ (Therborn, 2011:59), but this encounter was materialised in many different ways and mutually affected each party in the relationship.

Additionally, the apparatus of colonial-imperialism was not a homogenous, unstoppable behemoth but an assemblage of diverse, and often conflicting, institutions and interests (Comaroff, 2002:130; Nuttall, 2009:2-3). Thus, as Therborn (2003:297) has recognised, entanglement also captures the open-ended, indeterminate dimensions of colonial-imperialism.

This type of entanglement also includes the historical entanglement of Rwanda with

Burundi, and of both with the territories in the wider Great Lakes region. This category necessarily entails a consideration of intensity. Societal entanglements are not constant over time; we might say that they ‘tighten’ and ‘loosen’ at specific historical junctures.

For instance, Burundi and Rwanda were administered under indirect colonial rule as

Ruanda-Urundi, demonstrating a much stronger interrelationship with each other and with European states like Belgium than that which has been the case in the post- independence histories of the two countries. Furthermore, one can distinguish vectors of entanglement which are more or less intense at specific junctures. For instance, that of ethnic conflict, its occurrence in one country and potential to significantly influence conflict in the other, has been an especially significant entanglement in the post- independence era.

3. Geohistorical entanglements

These are the entanglements of modern societies with what Arnason (2003a:308; see

also Nuttall, 2009) calls ‘a global geo-economic, geo-political and geo-cultural

constellation which changes in the course of Western expansion and the unfolding

responses to it’. They involve entanglements with institutional structures and processes

of transformation, external mechanisms from which specific trajectories of modernity

derive influence and to which they have to adjust. These include cultural

entanglements: in this case, for example, we might think of those intertwined with the

influence of the Catholic Church and, later, humanitarian and development NGOs.

There are also political entanglements entailed in relationships with institutions such as

the League of Nations, the United Nations, and the influence of the Cold War blocs and

their competing ideologies. There are legal entanglements bound up with changing

global legislation on human rights and crimes against humanity. Finally, there are

entanglements that are principally economic in nature; in Burundi and Rwanda’s case,

we can include the changing institutions of economic and resource appropriation under

colonial rule, relationships of aid donorship, and later institutions of neoliberal

capitalism like the International Monetary Fund and the World Bank and their structural

adjustment programmes.

THE ROUTES TO AND THROUGH MODERNITY Furthermore, the framework presented here conceptualises modernity processually, rather than as an achieved state or condition. Plurality is not solely located in spatially distinct units, but also in transformations in modernity – defined globally and in the singular – over time (Arnason, 2005; Wagner, 2010). Trajectories of modernity unfold along specific and open-ended routes which are path-dependent but not teleologically determined. Therborn

2010:73) identifies ‘four major pathways into modernity’, each defined by ‘the conflict lines 76

for and against the new, between modernity and tradition, between modernity and anti- modernity’. These are:

1. The internalist European route, which emerged not as a natural emanation of European

civilisation, but ‘out of conflicts internal to Europe, to North-Western Europe

primarily’, and revolutions which ‘pitted the forces of reason, enlightenment,

nation/people, innovation, and change against those of the eternal truths of the Church,

of the sublime wisdom and beauty of Ancient philosophy and art, of the divine rights

of kings, of the ancient privileges of aristocracy, and of the customs of fathers and

grandfathers’ (Therborn, 2003:299, 2011:56-57).

2. The route of New World modernity, including the USA and other societies of the

Americas, Australia, South Africa, etc. These involved the mobilisation of ideals and

models supplied by the Enlightenment and carried by European settlers; Louis Hartz

(1969) referred to them as ‘fragment’ societies, to describe their ‘broken off’ character

from Europe. Yet indigenous populations, slaves and other racialised figures were

excluded from these Enlightenment ideas. As such, ‘New World modernity was a

modernity of settlers and slave plantation owners, with downtrodden moderns nearby,

the natives, the slaves and the ex-slaves. Traditional authority was the colonial

metropolis and its governing representatives sent from overseas… Nevertheless, for all

their imperial loyalty, the white Dominions set out on their own settler course to

modernity, quite distinctive from that of the motherland’ (Therborn, 2011:58).

3. The third route, the most significant for present purposes, was the colonial route. As

Therborn (1995:6) puts it:

To the Colonial Zone modernity arrived from the outside, literally out of the barrel of guns, while resistance to modernity was domestic, and crushed. Later on, colonial modernity involved the acculturation of part of the colonised, their learning the appropriate ideas of the colonisers – popular sovereignty, national self-determination, socio-economic development – and their turning them against the coloniser and the 77

metropolitan masters. From this followed, inter alia, a deep cultural trauma – with a potential for extraordinarily creative new combinations to the extent that the trauma can be mastered – social fragmentisation and the primacy of the national issue. The colonial route to and through modernity was thus in almost all cases succeeded by

that of anti-colonial rebellion (Therborn, 2010:75). As applied to the specifics of the

cases of Burundi and Rwanda, I prefer to say that the colonial route was succeeded by

the decolonial route. I will discuss this terminological choice below.

4. Finally, a fourth route38 is located in the adaptive and reactive modernisation exhibited

in countries like Japan and Turkey. This was ‘externally induced modernisation,

challenged and threatened by the new imperial powers of Europe and America, where

a section of the ruling elite selectively imported features of the threatening polities in

order to stave off colonial subjugation’ (Therborn, 1995:6; see also Kaya, 2004:7).

I will now locate Burundi and Rwanda into the typological framework outlined above. This constitutes the organisation of the substantive empirical chapters, so I will also use this as an opportunity to discuss some of the themes to be addressed.

THE COLONIAL ROUTE In chapter six, I use Therborn’s (2003) framework to discuss the ‘colonial route’ to and through modernity. In the ‘colonial zone, from North-western Africa to Southeast Asia’, modernity arrived not only ‘out of the barrel of guns’, but from the teachings of missionaries and the arrival of medical experts, racial scientists, and so on. It was imposed from the outside, but to leave it here would be unacceptably to reduce the complexity of the encounter. As we

38 A fifth type is proposed in order to classify those ‘hybrid routes’ which exhibit elements of more than one route. ‘The four routes are actually existing historical trajectories. But they may also be treated as Weberian ideal types, so that concrete historical experiences might include aspects of two or more ideal routes’ (Therborn, 1995:7). Included here are Russia, containing both reactive/adaptive strands and European strands. Protectorates such as Egypt and Morocco combine elements of the colonial route and the reactive/adaptive route. 78

shall see, colonial powers became entangled with the societies that they encountered on the ground. These societies, as I demonstrate in chapter five, were dynamic formations developing of their own accord. Thus, colonial modernity has important, distinctive characteristics. As

Arnason (2003:342) puts it:

The question of colonial modernity is of major importance, both in its own right and because of the interconnections between colonial and metropolitan developments. To begin with general considerations, colonial modernity as such has distinctive characteristics due to its cross- societal and cross-cultural context. The modernising structures and strategies at work in colonial settings differ from those of metropolitan societies because of their direct alignment with imperial power; their concrete social frameworks are shaped by interaction with indigenous traditions and power structures; finally, the multiple forms of active and passive resistance affect the trajectories of colonial rule. My principal argument is that colonial modernity in its indirect form had a paradoxical effect on the relationship between modernity and tradition. As Therborn (2010:74) points out,

‘European colonial modernity was in fact a mixture of a very delimited modernist thrust and neotraditionalism – of bolstered ‘indirect rule’ and codified ethnic identities and customary law’. This is particularly significant in the context of the assumptions of earlier modernisation theory, some of which are replicated in decolonial and postcolonial theory, and which are revitalised in the detraditionalisation thesis of scholars like Giddens and Beck. Here, as we have seen, modernity is said to involve an orientation away from the past, or tradition, towards an open-ended future. Multiple modernities, as discussed, constitutes an important advance inasmuch as it highlights the existence of tradition in any variant of modernity. However, what has been underemphasised, resulting from a lack of engagement with colonial modernity, is the invention of tradition and its racialisation: the colonisation of the ‘cultural programme’ of the colonised (Said, 1978; Hobsbawm and Ranger, 1983; Ranger, 1983; Mudimbe, 1988)

THE DECOLONIAL ROUTE An autonomous interpretation and institution of modernity emerged as a distinct possibility in the 1950s. Burundi and Rwanda in this period ought to be situated as entangled in the structural problems inherent in the global transformation of modernity, of which 79

decolonisation is an expression. Modernity was thus carried further by ‘new generations of natives, of évolués who turned what they had learnt from their conquerors – about the possibility of change and development, about nations, peoples, rights – against their masters and created anti-colonial nationalism. This is the road to modernity by anti-colonial rebellion’

(Therborn, 2010:74). ‘Anti-colonial rebellion’ is a misleading title, so I will opt for the more neutral ‘decolonial route’, suggesting the route to and through modernity through the formal withdrawal of colonial administrations. As we shall see, in Rwanda the dynamic that drove the independence movement and the social revolution of 1959-62 directed anti-colonial sentiments towards the Tutsi monarchy, and towards a wider substratum of society deemed representative of it. This occurred with the support of the departing colonial powers. Burundian independence was a more straightforwardly ‘anti-colonial rebellion’, but here still there were a number of conflictual interpretations of the legacy of colonial rule which undermine the homogeneity implied within the rubric ‘anti-colonial’. I will look principally at the formation of political parties and the tenor of elite social discourses in both countries.

THE POSTCOLONIAL ROUTE ‘Analyses of colonial modernity’, Arnason (2003:342) suggests, ‘must also deal with a closely related issue: the perpetuation of its distinctive pattern – in more or less modified form

– after the end of colonial rule. But the transition from colonial to post-colonial conditions is indisputably a turning-point which opens up new paths for historical interpretation’. If chapter seven concerns the ‘transition which opens up new paths for historical interpretation’, chapter eight concerns the ‘perpetuation of the distinctive pattern’ of colonial modernity following the formal process of decolonisation. I claim that the extreme violence which has indelibly marked the post-independence histories of Burundi and Rwanda, manifested in the 1972 genocide of

Hutu in Burundi and the 1994 genocide of Tutsi in Rwanda, can be situated appropriately in the context of the tension between modern imaginary significations which emerged in the 80

transition to independence and the racialised framework of tradition born of the colonial entanglement. This tension played out against a background of severe economic and political crisis, dependence and immiseration.

Genocidal violence, I claim, has been employed by elites to effect mastery over these crises and restructure social relations in order to entrench their power over the postcolonial state. Beyond this, these endeavours have found significant support from broader sections of the population, humiliated and frustrated by the disappointments of postcolonial modernity, whose collective ethnic identities are narrated with regard to distinct historical events and experiences of perceived injustice. Genocide has been a mass societal and communal mobilisation that cannot be disentangled from modernity. In this sense, genocides in Burundi and Rwanda are emblematic of what Eisenstadt called ‘traumas of modernity’ (2000:25). The conceptualisation of modern genocide should not be derived from some systemic or institutional feature of modernisation such as bureaucratic rationalisation or modern science; it ought instead to be seen as the intentional action of interpretive agents situated and constrained in a field of tensions, and to be situated in a distinct historical route through modernity.

To summarise, I conceive of modernity first and foremost as a time-orientation accompanied by a distinct conception of human agency, whereby the future becomes an open space to be shaped according to the possibilities within the present. It is most manifest in the emergence of a sensibility or condition whereby ‘human beings think of themselves as setting their own rules and law for their relation to nature, for their living together, and for understanding themselves’ (Wagner, 2001:38). This time-orientation is manifested in particular institutions and practices, related to ‘cognition, art, economics, and politics’

(Therborn, 2010:72). It is expressed in imaginary significations like autonomy, social justice, progress, liberation, and so on. These imaginary significations are situated in relation to each other within a field of tensions containing inherent contradictions and antinomies. This leads 81

to conflicting interpretations, and produces seemingly paradoxical social forms and practices.

These interpretations are situated within the boundaries formed by particular historical experiences and socio-cultural traditions. This variety of interpretations results in distinct institutional orders: this is the multiplicity of modernity.

These orders, however, do not exist in a state of separation, nor are their origins reducible to endogenous processes. Instead, there is a singular global modernity with distinct routes which are entangled with each other. These entanglements entail asymmetrical power relations and have historically been accompanied by processes of colonialisation and racialisation. Such asymmetrical relationships of power severely constrain the capacity of modern subjects in postcolonial contexts to act upon the interpretation of the imaginary significations of modernity. The working out or interpretation of these contradictions and antinomies can be exceptionally violent and traumatic. More specifically, working through these contradictions can in some cases involve genocide, when genocide is understood as a distinct form of reconfigurative or creative social practice that emerges in times of acute crisis such as wars, revolutions, coups or economic crisis. Genocide, in this sense, is related to modernity when conceived predominantly as a time-orientation – as is genocide prevention.

Before I move onto the substantive core of the thesis, where I tackle these issues in a sustained engagement with the historical route to and through modernity of Burundi and

Rwanda, I will use the following chapter to explain my methodological choices.

CHAPTER FOUR – UNDERSTANDING ENTANGLEMENTS: HERMENEUTICS, HISTORICAL SOCIOLOGY, AND METHOD

I have suggested that the best way of thinking about the Burundian and Rwandan experiences of the twentieth century is to conceive of them as entangled historical routes to and through modernity. In the sense that modernity conceived on this basis is a global phenomenon

(not a universal one)39 with multiple and specific articulations, ‘it should be the study object of a global history and a global social science’ which entails ‘focusing on global variability, global connectivity, and global inter-communication. It also implies a global look at processes of change, of continuity and discontinuity’ (Therborn, 2003:295). Therborn also suggests that

‘questions about how the contemporary world has been shaped are issues of self-conception and self-understanding, as well as of genealogy and historiography’ (Therborn, 2003:294). I suggest that this can apply to the self-understandings of disciplines like sociology and to their organising concepts such as modernity. There is therefore another advantage to the framework of entanglement, namely that it has the potential to be the basis of a reflexive historical sociology, to use Arpad Szakolczai’s term (2000).

The thesis can thus make a distinct contribution to Saïd Amir Arjomand’s recent programmatic call for an increased sociological focus on the potential contribution of regional and area studies40 to the evaluation of key concepts. In his words (2014:3):

The rich stock of concepts and theories that are mainly embedded in Western historical experience can be modified… through their dialogical engagement with concepts which are at

39 As discussed in the previous chapter, this distinction is proposed in contrast to the Eurocentric conception of European modernity as a universal phenomenon that was formed endogenously and provided an institutional and normative developmental model for the rest of the world. Global modernity, by contrast, is marked by a multiplicity of specific, entangled historical trajectories whose overall development cannot be subsumed under teleological conceptions of progress or regress (see also Bauman, 1998:59). 40 By area and regional studies I mean those bodies of literature which collect localised studies produced by scholars across the social sciences and humanities in disciplines such as anthropology, history and political science. These are oriented towards generating deep knowledge of the particularities of different cultures, histories and languages (Schäfer, 2014:146). They include African Studies, which can be subdivided into distinct regional specialisms (i.e. Central Africa, Western Africa, and so on).

last being formed on the basis of the vast, understudied, and analytically untapped historical and cultural experience of other regions and civilisations. Hence, the promise of comparative sociology for our generation, and of the present venture to realise this promise by integrating the findings of regional studies into social theory. The present work seeks to understand how a set of related events and processes – colonialism, independence and genocidal violence in Burundi and Rwanda – which have tended to remain neglected or defined as specialist pursuits in sub-disciplines, might aid the expansion of a central category in sociology (modernity). It does so in the light of recent and contemporary deconstructive criticisms and amid continuing calls for a ‘global sociology’.

In this chapter, I argue that philosophical hermeneutics constitutes a thorough grounding for this project in the philosophy of social science. This argument unfolds alongside an affirmative discussion of the interdisciplinary integration of social theory and area studies, outlined by Arjomand. I suggest that any such project must proceed in the manner of historical sociology. I then move on to a more practical discussion of how these methodological issues impact on this particular study, before outlining the methods employed specifically in the case study chapters.

PROBLEMS OF GLOBAL AND HISTORICAL UNDERSTANDING

At this preliminary stage, however, I want to situate this in a contemporary tension which affects endeavours towards inter-cultural/-historical understanding. For analytical purposes, this tension can be said to have two axes, one geographical and one historical.

Geographically, it consists in the tension between processes and institutional structures of globalisation and the situated location of knowledge-claims. The historical tension is located at the meeting point between, on the one hand, a longstanding recognition of the importance of long-term historical processes for the shape of any given society at any given time, and on the other hand, contemporary institutional and political tendencies toward a sociological presentism or ahistoricism. 84

There is widespread agreement in the discipline of sociology that contemporary societies are converging in many ways due to processes of globalisation and transnationalisation, including time-space compression, and flows of culture, capital, and people, even if there are disagreements concerning the periodisation, direction and effects of these processes41. It is said that the planetary condition is constituted by intertwining and overlapping ‘communities of fate’ (Held and McGrew, 2003:39) where events on one side of the world can have significant consequences for the other. A sociological response to this condition has been a sustained attempt to transcend what has been called ‘methodological nationalism’; the analysis of social processes, it is argued, can no longer be conceived as bounded within the container of the nation-state (e.g. Wimmer and Glick-Schiller, 2002; Beck,

2007; Amelina, 2012; Chernilo, 2011). The most pressing contemporary challenges – such as those related to the climate, to the economy, to security – are global in nature and present themselves to humanity, even if its effects are distributed differentially (Beck, 2009).

At the same time, important insights from feminist and postcolonial theorists, influenced by poststructuralist epistemologies, have stressed the importance of ‘standpoint’, locality and identity in the production of knowledge (Harding, 2004; Connell, 2007; Shilliam,

2011). These have highlighted the role of knowledge as a distinct form of capital which produces power effects and reinforces inequality. It is argued that much social thought from its classical origins is suffused with ethnocentric, patriarchal and outright chauvinistic assumptions veiled by claims to scientific detachment and value-neutrality. The overall effect, as Fuyuki Kurasawa (2000:12) has highlighted, is a paradox whereby:

On one hand, in light of the globalizing tendencies that are reshaping the world’s socio-political and economic landscape, the need for cross-cultural research has never been greater. On the other hand, the perils of such endeavours have been accentuated by the so-called ‘crisis of

41 This is demonstrated by the prevalence of the appearance of ‘global’ as an affixing qualifier in the title of many introductions to sociology (e.g. Held and McGrew, 2003; Macionis and Plummer, 2008; Cohen and Kennedy, 2013). 85

representation’ in the human sciences, itself prompted by the perceived ethnocentric bias of most Western approaches to the study of non-Western societies and cultural contexts. A similar problem affects the historical imagination of sociology. Elias (1987) detected a ‘retreat of sociologists into the present’ a few years before the fall of the Berlin Wall. He referred to the influence of the clash of value-orientations of Marxist and functionalist sociologists in the Cold War context and as such the summation cannot be sustained as a critique of the contemporary state of the discipline. Indeed, the paradigms of multiple modernities and civilisational analysis emerged as serious historical-sociological projects partly as a response to the triumphalist and essentialist discourses of the ‘end of history’

(Fukuyama, 1989) following the collapse of the Soviet Union. Yet it has been claimed that a presentist orientation particularly marks contemporary British sociology; it is suggested that contemporary institutional conditions and transformations, funding directives and governmental pressures work against the study of long-term historical processes (Inglis, 2013).

At a time when sociological research output continues to be measured in terms of its impact

(Holmwood, 2010), it has therefore to justify its existence and continued funding in terms of its adequacy to respond to present-day policy concerns. The relative decline of a historical orientation to the study of social processes (Abrams, 1982:2), beyond its professionalisation in the subdiscipline of historical sociology, might also relate to contemporary political and value- orientations in the context of struggles in the present and the compulsion to delineate future alternatives (Dawson, 2016).

This jars with the foundational sociological premise that any human society is a crystallisation of long-term historical dynamics and unplanned processes beyond the boundaries of the nation-state that render human beings interdependent and constrains their capacity to act freely. As Marx (1852) most succinctly summarised, human beings make history but they do not do so under conditions of their own choosing. Thus, to think sociologically is to think historically (Abrams, 1980, 1982; Skocpol, 1984; Mills, 2000 [1959]). This extends to 86

the argument made forcefully by Gurminder Bhambra (2007, 2014) about the formation of the discipline of sociology, the occlusion of colonial relationships and the necessity of therefore

‘reconstructing’ its organising concepts. To ‘decolonise sociology’ entails a deep engagement with history (Rodríguez, Boatcă and Costa, 2010), but such a project of decolonisation exists in tension with the ‘colonisation’ of the university by the ‘systemic’ imperatives (Habermas,

1987) outlined above which render this historical work more and more difficult.

This difficulty is compounded by virtue of the fact that problems similar to those of inter-cultural understanding also occur in the interpretation of societies at different points in history. If it is the case that ‘every present society has grown out of earlier societies and points beyond itself to a diversity of possible futures’ (Elias, 1987:226), it follows that the ‘inner lives’ of members of past societies are not equivalent to those in the present (Harrington, 2001:28).

We cannot, then, simply ‘jump out of our skins’ into worldviews of other societies and historical periods and the thoughts of those who dwelled in them (Bauman, 1978; Bernstein,

1983:89; Harrington, 2016:53). In the sense that modernity is comprised of specific articulations resultant from entangled historical routes, the position of the researcher is also one of being entangled in the world. Nevertheless, as I will argue here, this situation of entanglement is not a windowless imprisonment within time and space but a horizon which opens on to the world and which can be broadened in the process of a movement from one’s

‘inherited gravity-point of perception’ (Harrington, 2016:53). I will now elucidate this in relation to the tradition of philosophical hermeneutics.

PHILOSOPHICAL HERMENEUTICS, HISTORICAL SOCIOLOGY AND AREA STUDIES

Hermeneutics refers broadly to the art, science or technique of the interpretation of written texts, and is especially concerned with how to approach and handle those written in times and places different from one’s own (Grondin, 1991:1; Outhwaite, 2007:459; Thiselton, 87

2009:1). Though hermeneutical practice has roots in antiquity42 and in traditions of theological, philological and juridical interpretation, it was systematised principally under the stewardship of Friedrich Schleiermacher in German-speaking Europe during the nineteenth century. It was he who established the term verstehen or understanding as central to the interpretation of the meaning of texts and to reconstructing the intentions of authors (Bleicher, 1980:14-15;

Outhwaite, 2007:460). Hermeneutics was then developed by figures like Giambattista Vico and Johann Gustav Droysen in terms of providing an historical understanding of the place of distinct individuals, events and cultures in world-historical totality (Bleicher, 1980:17). Later,

Wilhem Dilthey developed hermeneutics into a foundation for the specifically human sciences

(Geisteswissenschaften) and established ‘what we would now call a research programme for history and the other human sciences based on the distinctiveness of human psychic expressions and the understanding of those expressions’ (Outhwaite, 2007:60; see also Burke,

1993:8).

As is well documented, the interpretive tradition in sociology developed out of this context (Bleicher, 1980:25-26; Harrington, 2001:2; Poggi, 2005). It was pioneered principally in the work of Max Weber and his attempt to combine the ‘interpretive understanding of human action’ with the ‘causal explanation of its course and consequences’ (Weber, 1978:4). His influence is strong in the work of the theorists of the multiplicity of modernity introduced in the previous chapter. Simmel’s sociology of the forms of interaction represented a divergent movement of interpretive sociology, whose influence can be seen in the later programmes of symbolic interactionism and ethnomethodology (Poggi, 2005). Interpretive sociology thus ties together a multiplicity of research paradigms. The common unifying thread is that human

42 Hermeneutics derives its name from Hermes the interpreter and messenger, the Greek God of transition and boundaries (Grondin, 1994:22; Gadamer, 2004 [1975]:165; Diggs, 2010). Hermes is a complex and ambivalent character, presented in some mythologies as a thief or trickster (see Kerényi, 2015 [1944]; Szakolczai and Wydra, 2006:140), thus perhaps serving as a warning of the ‘objectifying’ or ‘dispossessing’ potential even in hermeneutics. 88

beings, individually and collectively, are meaning-making and interpretive animals (Poggi,

2005:65). Thus, the sociologist cannot simply explain phenomena in the manner of a natural scientist; rather they must attempt to understand the intentions and motivations which underpin social action.

Paralleling the trajectory of interpretive sociology during the twentieth century is that of the tradition of philosophical hermeneutics. Similarly responsive to the work of Dilthey on the specificity of the human sciences, it also has roots in Edmund Husserl’s phenomenology before being elaborated most systematically in the works of Martin Heidegger and Hans-Georg

Gadamer (Grondin, 1991; Moran, 2000; Schmidt, 2006). In the following, I especially focus on the work of Hans-Georg Gadamer as presented in his magnum opus Truth and Method

(2004 [1975]). I will identify and outline several points, supplemented by more sociologically- oriented thinkers, which can help to overcome the obstacles to understanding outlined above.

This will lead into an explicit claim that philosophical hermeneutics provides a thorough grounding in the philosophy of social science to contemporary endeavours to integrate the findings of area/regional studies with the conceptual framework of sociological theory. I suggest that it contributes in three ways. Firstly, it presents a systematic way of approaching and understanding texts from specialist literatures in the manner of an interdisciplinary dialogue with social theory. Secondly, particularly in the ontological turn of Heidegger and

Gadamer, philosophical hermeneutics opens up action and events situated in distinct historical periods and cultural regions to understanding. Thirdly, as I discussed in the previous chapter43, there is a hermeneutic dimension to the multiplicity of modernity and societal self- understandings that opens them up to sociological interpretation.

43 See pages 55-66. 89

A fundamental principle of philosophical hermeneutics, particularly as represented by

Heidegger and Gadamer, is that interpretation does not merely demarcate an approach to method in the human sciences but is rather the distinctive character of human ‘Being-in-the- world’ as such. Human beings are beings for whom Being is meaningful (Polt, 1999:29-30).

Insofar as hermeneutics constitutes a methodological basis for the human sciences, this is only a radicalisation of and systematic reflection on a universal principle at play whenever we understand anything (Gadamer, 2004 [1975]:267; see also Bernstein, 1983:34; Outhwaite,

1987:62; Polt, 1999:68-74; Harrington, 2001:24; Moran, 2000:234-238; Inwood, 2000:22).

Understanding occurs from a distinct standpoint or ‘thrown-ness’ in the world against a background shaped by one’s contingent locatedness and acculturation in space and time (Polt,

1999:46). Human beings are ‘‘thrown’ into the world as beings who understand and interpret’

(Bernstein, 1983:113; see also Moran, 2000:13). As I have said, it is widely accepted that we live in social spaces embedded within broader global processes, where time and space has been compressed, and the significance of events is felt in different locales simultaneously. It follows that understanding is not only possible of different historical periods and cultural regions and their embeddedness in a connected world, but that it is a necessity for any avowedly global sociology. However, our capacity to understand is filtered through preconceptions of the world created in large part by the accident of our ‘thrownness’ in situated locations, historical and geographical, and the resources that are afforded us therewith. These locations are the crystallisation of unplanned economic, political and social processes, such as state-formation, geopolitical power struggles, colonial-imperialism and capitalist expansion. As such, we understand the world and our position in it within specific constellations of asymmetrical power relations.

In this sense, it might be said that the traditional social-scientific concepts that emerged in the West are suffused with prejudice. We have seen in the previous chapter how the concept 90

of modernity has often been seen as such, especially in the form of postcolonial and de-colonial critique 44 . However, Gadamer sought to rehabilitate prejudice, or ‘pre-judgements’. The

Enlightenment, he argued, had fostered a ‘prejudice against prejudice’: ‘The history of ideas shows that not until the Enlightenment does the concept of prejudice acquire the negative connotation familiar today. Actually ‘prejudice’ means a judgement that is rendered before all the elements that determine a situation have been finally examined’ (Gadamer, 2004

[1975]:270). Contained in the prejudice against prejudice is the erroneous assumption that it is both possible and desirable to break from inherited standpoints; to step outside history. As

Gadamer (2004 [1975]:276-277) puts it:

History does not belong to us; we belong to it. Long before we understand ourselves through the process of self-examination, we understand ourselves in a self-evident way in the family, society, and state in which we live. The focus of subjectivity is a distorting mirror. The self- awareness of the individual is only a flickering in the closed circuits of historical life. That is why the prejudices of the individual, far more than his judgements, constitute the historical reality of his being (see also Ricoeur, 1991:72; Hammersley, 2011:135). Nevertheless, understanding is not fatally curtailed by the incommensurability of our prejudgements with the object to be understood; understanding is not a vicious circle. Key here is the idea of the ‘hermeneutic circle’. The hermeneutic circle holds that ‘any interpretation which is to contribute understanding, must already have understood what is to be interpreted’

(Heidegger, 1996 [1953]:142; see also Ricoeur, 1967:352; Polt, 1999:69; Moran, 2000:235;

Simms, 2003:37; Outhwaite, 2008a:23). Any scientific investigation always proceeds from our everyday, ordinary encounters with the phenomena under investigation 45 . We approach phenomena with questions shaped by inherited ‘prejudices’ or ‘fore-understandings’

(Heidegger, 1996 [1953]:141) – drawn from both our ‘pre-rational familiarity with the world and… the millennia of cultural and philosophical tradition’ (Polt, 1999:67) – and we modify

44 See pages 52-55 45 This puts the autobiographical section of my introductory chapter in perspective. For instance, the BBC television report that I mentioned (see pages 5-6) constitutes an obstacle to understanding only if the prejudices which underpinned the encounter had not been questioned. Both that encounter and this project can be seen as nodal points in the hermeneutic, spiralling movement of understanding. 91

these understandings in the manner of a dialogue ‘during the ongoing process of research, understood as a dialogical encounter with otherness that propels us into self-awareness’

(Harrington, 2001:30; see also Gadamer, 2008 [1976]:38;). Our pre-conceptions of the world are not obstacles in the way of objective understanding. Rather, by making phenomena comprehensible to us, they are what make understanding possible. They are vital tools that are projected onto what is to be understood and which must keep being reworked in the confrontation with the world (Kögler, 2006:204). In Gadamer’s (2004 [1975]:267) words:

‘interpretation begins with fore-conceptions that are replaced by more suitable ones… Working out appropriate projections, anticipatory in nature, to be confirmed ‘by the things’ themselves, is the constant task of understanding’.

This dialogical process of understanding a text, a historical phenomenon or another social actor constitutes, in Gadamer’s evocative terms, a ‘fusion of horizons’ (Gadamer, 2004

[1975]:306, 2006:45; Harrington, 2001:30). ‘Horizonality’ has important precursors in the phenomenological and hermeneutic traditions46. For Gadamer, a horizon refers to the whole of our background of experience structured by inherited preconceptions and socio-cultural traditions that make understanding possible. Like visual horizons, an epistemic horizon refers to a ‘range of vision that includes everything that can be seen from a particular vantage point’

(Gadamer, 2004 [1975]:302; see also Schmidt, 2006:105; Nenon, 2016:248). These horizons are not insurmountable but move ‘with one and invites one to advance further’ (Gadamer, 2004

[1975]:245). Furthermore, horizons are not distinct and incommensurable with one another.

Understanding spatially and historically differentiated forms of life does not proceed ‘by

46 These include: Dilthey’s argument that individual human lives do not unfold as a series of isolated events but are instead processes set against the background of previous experiences, which in turn anticipate future experiences; Husserl’s concept of the lifeworld (lebenswelt) of pre-theoretical experience and its historical and cultural horizons, as well as the horizons of the ‘intentional objects’ of perception; and Heidegger’s notion of Dasein and its situatedness in commonly understood, historically developed understandings (Grondin, 1994:92-96; Gadamer, 2004 [1975]:245; Moran, 2000: 161, 181- 186; Nenon, 2016). 92

immersing oneself in their specific uniqueness or re-living them as if ‘from inside’, but by following an exactly opposite strategy: by spotting the general in the particular, by enlarging both the alien and one’s own experience so as to construct a larger system in which each ‘makes sense’ to the other’ (Bauman, 1978:218). Through its fusion with those of others, then, our own horizon can thus be broadened; to possess and recognise a horizon means that ‘one learns to look beyond what is close at hand – not in order to look away from it but to see it better, within a larger whole and in truer proportion’ (Gadamer, 2004 [1975]:305; Nenon, 2016:251).

Furthermore, beyond being shaped by historical traditions and by our everyday experience of interpreting the world, our horizons are also formed in the context of our position in social relations. As both Heidegger and Gadamer emphasised, Being-in-the-world always entails being with others (Polt, 1999:56, 60-64; Gadamer, 2004 [1975]:304). Here, philosophical hermeneutics reveals an affinity with some currents in twentieth century sociological thought. For Alfred Schutz (1967), for instance, who pioneered the phenomenological approach to sociology, meaning is not constructed subjectively in the individual mind but is rather produced and shared intersubjectively through social interaction over time (Harrington, 2000). Elias (1978:1-32, 125-133) also positioned his own approach against the ‘egocentric’ conception of individuals as closed entities, or in his words homo clausus (closed human), standing in relation to the similarly reified concept of society. Rather, human beings are homines aperti (open humans) and are always situated in relation to other human beings in complex ‘figurations’, or indeterminate relations of interdependence. Not only is the ‘habitus’ of the individual formed in relation to those of others; the habitus involves elements that are shared (Elias, 2010 [1985]:108). Thus horizons are invested with an intersubjective capacity resultant from human interdependence. It is in this sense that we can speak of the horizons of ‘societal self-understandings’ (Wagner, 2010a; Wagner and Friese,

2000). 93

Elias (1978:131) stressed that interdependencies of homines aperti form ‘a village, a city or a nation’ whose status as ‘figurations cannot be perceived directly’. Furthermore, he recognised that the ‘ethnocentric image of humanity divided into national states’ was analogous to the individual model of homo clausus and ought to be revised in favour of a ‘conception of one’s own nation as one among many other interdependent ones’ (Elias, 1978:29-30). His neglected contributions to the sociology of international relations were made within the context of Cold War hostilities (Elias, 2010 [1985]); it is open to question how he would have conceptualised interdependence in terms of more contemporary debates concerning globalisation. For Bauman (2008:73), the condition of interdependence has reached a planetary level47, even if an awareness of this interdependence is yet to emerge:

The globalization process has thus produced a network of interdependence penetrating every nook and cranny of the globe, but little else. It would be grossly premature to speak of even a global society or global culture, not to mention a global polity or law… The planetary reach of capital, finances, and trade – the forces that decide the range of choices and the effectiveness of human action, the way humans live and the limits of their dreams and hopes – has not been matched on a similar scale by the resources that humanity developed to control those forces that control human lives. Bhambra (2014) goes further than Bauman to speak of ‘connected histories’, following

Sanjay Subrahmanyam. The contemporary planetary condition of interdependence is attributable to the historical interconnections between European and Western centres of power and the rest of the world. Globality is not a recent phenomenon in modernity; modernity was

‘always-already global’ (Bhambra, 2014:ch. 7). Therborn (2014:267) also notes the role of historical and geographical connectedness in the formation of the contemporary world and at

47 One has to be careful not to overstate this global interdependence or view it teleologically. In 2016, at the present time of writing, right-wing populism with a strong nationalist and isolationist anti- globalisation component is making significant gains across the northern hemisphere, including but not limited to: the vote to leave the European Union in the UK; the election of Donald Trump to president of the USA; and significant electoral gains of Far Right parties across Western and Eastern Europe. Thus, any claim about planetary interdependence ought to take seriously the prospects for reversals of processes and institutions of global connection, in the same sense that Elias (1994 [1939], 2010 [1985], 2013 [1989]; see also Fletcher, 1997) argued that the processes of civilisation are not unidirectional but are instead susceptible to reversal in the form of decivilising processes. 94

the same time addresses the importance of relativising our horizons in light of our spatially and temporally specific condition of being entangled in the world:

The part, the region, of the world we are living in today was molded long before yesterday. This historical background shapes (virtually) all of us, albeit not in any unilinear, deterministic sense, by outlining our cognitive horizons and our probable paths of learning, by providing bearings of our transcendental and moral convictions, by channelling our sexuality, by filling our memory, by guiding our tastes, by forming our sociability, even by supplying our sense of ourselves. A major task of social science, then, is to excavate this sociocultural geology of our lives. As such, for global sociology to exist, it has to be both possible and necessary to speak to experiences situated within other cultures and historical periods. ‘Speaking to’ is not equivalent to ‘speaking for’ or ‘on the part of’ those experiences and silencing or extinguishing otherness through the scholar’s objectification of the other (Spivak, 1988). Rather, projections of scholars ‘first call otherness into being, into meaningfulness… and projections never need be in any way final: they always remain open to critical adjustment according to the specific, distinctive, previously unnoticed ‘parts’ of the ‘whole’ of the object in question that come into view through experience, through ongoing phenomenal encounter with an object’ (Harrington,

2016:54). All standpoints are thus cosmopolitan standpoints, located in and delimited by their entanglement in historical and geographical stratifications of power, but whose horizons open up to the world.

Philosophical hermeneutics offers the most plausible way of acknowledging the entanglement of expertise and scientific authority in technocratic and objectifying domination, including over colonial subjects and people today in the Global South, without retreating into a relativistic and descriptive auto-ethnography (Bernstein, 1983; Taylor, 2002:126). In its emphasis on application, it also avoids a particular kind of meta-theoretical critique of sociology with an excessive concern with ‘its epistemological and ontological conditions of possibility’ (Wagner, 2001:41). As Harrington (2016:367 emphasis added) puts it: ‘after all necessary redress of inequalities of voice, substantive problems of empirical inquiry remain. 95

Scientific paradigms have to be not only critiqued and deconstructed: they also have to be put to use’. It is through a dialogical engagement with an independently existing phenomenon or external social reality that our preunderstandings and prejudgements can be broadened.

Understanding unfolds in practical application (Gadamer, 2004 [1975]:307-341; Bernstein,

1983).

In the Gadamerian sense that through a fusion of our horizons with those of another we come to a better understanding via a revision of our own prejudgements, it is through an understanding of non-European historical trajectories that we can come to an understanding of the fallibility of the prejudgements entailed in social and political theoretic concepts that universalise from the particularity of the European experience. Any normative postcolonial critique of the Eurocentricism of the sociological concept of modernity, however, must at the same time demonstrate an awareness of how the concept of modernity and previous theorisations of it enable such a critique to emerge in the first place (Joas, 2003:2). In other words, any sociologist’s critique of the tradition of sociology ought to acknowledge what

Gadamer called the ‘effective history’ (Wirkungsgeschichte) of the tradition working upon them in ways both conscious and unconscious. As he argues: ‘historically effected consciousness is an element in the act of understanding itself and, as we shall see, is already effectual in finding the right questions to ask’ (Gadamer, 2004 [1975]:300-301).

I suggest that this complements the historically-orientated postcolonial critiques of modernity, such as that offered by Mbembe (2001:9):

Dealing with African societies’ ‘historicity’ requires more than simply giving an account of what occurs on the continent itself at the interface between the working of internal forces and the working of international actors. It also presupposes a critical delving into Western history and the theories that claim to interpret it (see also Costa, 2007:5; Seth, 2016) Therefore, in this sense, to ask questions of Burundi and Rwanda in terms of applications and extensions of the concept of modernity at three distinct temporal points – colonialism, 96

decolonisation and postcolonialism – is simultaneously to ask questions of the discipline of sociology and the purported Eurocentricity of its operative and organisational concepts. In this sense, I emphasise coloniality and postcoloniality as themes of ‘comparative or reflexive historical sociology’ (Arnason, 2003:343; see also Bernstein, 1983:36; Kurasawa, 2000:18;

Szakolczai, 2000; Wagner, 2008:7-8; Hammersley, 2011:101-103; Thomas, 2011:737).

Such a project is therefore oriented to the narration of the history of modernity discussed in the previous chapter. Bhambra (2007, 2014) has argued that the commonly- relayed narrative of the endogenous emergence and development of European modernity – which groups together key constellational events and periods such as the Renaissance, the

Enlightenment, the French, industrial and scientific Revolutions and so on – has significantly downplayed the colonial-imperial relationships which connected European state formation to the rest of the world (see also Fanon, 1965; Said, 1978; Mudimbe, 1988; Mbembe, 2001;

Ndlovu-Gatsheni, 2013).

Attempts to reconstruct or extend the concept so as to incorporate hitherto ignored events and processes such as colonial-imperialism are complemented by the narrative hermeneutics of Paul Ricoeur. For Ricoeur, the past cannot be placed before us as an object or a series of chronological facts waiting to be uncovered. Such an approach is a disavowal of our

‘thrown-ness’ in history. The reality of the past is thus only accessible to us by way of mediation in the form of ‘traces’, constituted by ‘marks, inscriptions, documents, archives, and the monuments of all kinds that play the role of ‘facts’ for historical inquiry’ (Ricoeur,

1976:691; see also Scott, 1990:4). Writing history consists of the ‘reworking of these traces into a re-presentation of the past in our present’ (Simms, 2003:95).

More than a process of revealing, Ricoeur sees working through history as a process of narrative reconstruction. The historical work of reconstruction thus assumes a relationship of 97

‘standing-for (or ‘taking the place of)… a past that is abolished yet preserved in its traces’

(Ricoeur, 1988:100; Simms, 2003:95). Narratives are produced via the ‘emplotment’ of events and transformations in the flux of time. Time itself is possessed of three degrees of organisation: ‘within-time-ness’, ‘historicality’, and ‘deep temporality’ (Ricoeur, 1980:170-

171; White, 1984:27-28). Within-time-ness, closest to ordinary representations of time, is the time ‘in which events take place’, events which are ‘datable’, ‘public’ and ‘memorable’

(Ricoeur, 1980:170-171). This level of organisation concerns ‘our thrownness among things – which make the description of our temporality dependent on the description of the things of our concern’ (Ricoeur, 1980:172). At a deeper level time is ‘historicality’, implying a strong emphasis on the ‘weight of the past’ (Ricoeur, 1980:172; Clark, 2002:28). The third degree of organisation, ‘deep temporality’, the ‘plural unity of future, past, and present’ (Ricoeur,

1980:171), is that which functions in the narrative. This dimension allows historicality to be glimpsed through the ‘plotting’ of events or actions that occur ‘within-time’.

In this sense, the narrative transcends the chronicling of disparate events in a series by seeking to interpret their individual significance as part of a meaningful whole (Elliott, 2005:3).

In the context of this project, then, ‘plotted’ events analysed in the area studies literature will demonstrate the entangled historical routes to and through modernity, which were more or less prompted by the colonial encounter between Europe and the Great Lakes region, from the beginnings of the twentieth century until its end. In this sense, specific events and actions can be analysed and interpreted in terms of their meaning for the broader framework. The difficulty of addressing a complex history over such a long period is offset by the fact that its ‘historicality’ can be ‘grasped together’ through the plotting of seemingly ‘scattered events’ (Ricoeur,

1980:178). Events such as genocides, coups, revolutions and the emergence of social movements are always situated in more deep-seated historical processes, even if they are not reducible to them. As Fernand Braudel (1995:901) surmised: 98

Events are the ephemera of history; they pass across its stage like fireflies, hardly glimpsed before they settle back into darkness and as often as not into oblivion. Every event, however brief, has to be sure a contribution to make, lights up some dark corner or even some wide vista of history. Thus, ‘in the writing of the historical text, the aim in view should be to represent (human) events in such a way that their status as parts of meaningful wholes will be made manifest’

(White, 1984:26). Again, this is also redolent of the hermeneutic movement between the part represented by the particular event or phenomenon under investigation and the whole represented by the broader historical and global context of modernity and its constitutive entangled trajectories. This is consistent with the general aims of historical sociology, ‘to place a particular event or phenomena [or a series of events or phenomena] in a broader context, both historical and global’ (Delanty, 2013:4; see also Abrams, 1980:5; Bryant, 2000:489). This broader context is typically ‘the formation and transformation of modernity’ (Delanty and Isin,

2003:1). As such, historical sociology is a mediatory position between theoretical generality and historical particularity (Rueschemeyer, 1984; Burke, 1993:2), which I suggest brings the concerns of social theory and area studies into a shared horizon.

USING PRIMARY AND SECONDARY DOCUMENTARY SOURCES

I will be consulting both primary documentary sources and secondary sources drawn from the area studies literature. To clarify, following Gidley (2004:250), by ‘documentary sources’, I refer predominantly to ‘written sources… produced at the time of the events described, or some time later’. By primary sources, I refer to records that have survived from the time of the events described; secondary sources are those ‘accounts created by people writing at some distance in space or time from the events described’, which have already undertaken interpretations and constructions of various traces. This study draws principally on the latter. Taken together, these documents ‘provide multiple, prismatic vistas onto the conditions of life, the thoughts and actions, of those who participated in past social worlds that were no less real than our own’ (Bryant, 2000:499). 99

According to John Scott (1990:1), ‘the general principles involved in handling documents are no different from those involved in any other area of social research, but that the specific features of documentary sources do require the consideration of their distinguishing features and the particular techniques used to handle them’. He distinguishes four interrelated criteria for the adequate handling of historical documents, each of which I consider in the remainder of this chapter. These are: authenticity (concerning the document’s genuineness); credibility (concerning the extent to which the observation is accurate); representativeness

(concerning judgement as to whether the individual documents consulted are representative of the totality of relevant documents); and meaning (concerning the understanding of the meaning or significance of the document) (Scott, 1990:19-36; see also Platt, 1981; Mogalakwe, 2009:51-

53).

Primary documents from the colonial period, addressed in chapter five, include those in the Jean-Marie Derscheid online archive on Burundi and Rwanda, as well as official publications of the Belgian colonial administration consulted both online and at various locations such as the British Library. Documents pertaining to colonial exploration and early ethnographic description, as well as broader racial theory and other subjects, were consulted in physical form, microfilm and in online archives. For chapter six, which concerns independence movements, I consulted official publications of political parties, speeches and articles by intellectuals, in addition to government documents such as constitutions. On the subject of postcolonial violence, addressed in chapter seven, I have utilised sources from the Rwandan genocide archive, transcripts of radio broadcasts and pamphlets (translated into English and/or

French), as well as reports from international organisations such as the United Nations, Human

Rights Watch and Amnesty International.

With regard to the sampling and acquisition of secondary texts, I have made extensive use of the very substantial World Bibliographic Series editions on Burundi and Rwanda, in 100

addition to the online information presented on the Oxford Bibliographies webpage. The former are very wide-ranging, the latter are regularly updated, and both are compiled by area studies experts. The consultation of secondary texts has led to the discovery of other texts in the manner of snowball sampling. This is consistent with a hermeneutic approach in which it is permissible and even necessary for one to ‘engender and entertain hunches, explanatory propositions, ideas and theoretical elements’ (Friedrichs and Kratochwil, 2009:709, see also

Locke et al, 2008:908-9), an approach that the US pragmatist philosopher Charles Sanders

Peirce (1998 [1903]) called ‘abduction’.

The distinction between a primary and secondary source, it should be noted, is not always an easy one to make. For example, Thomas Patrick Melady’s Burundi: The Tragic

Years (1974) is an historical examination of the roots and causes of the 1972 genocide in

Burundi. It is, however, an eye-witness account based on his experiences as US ambassador to

Burundi at that time. Similarly, many accounts of colonial explorers were the after-the-fact recounting and rationalisations of their travels, which were likely to have been much more chaotic than their orderly and chauvinistic presentation suggests (see Fabian, 2000).

Furthermore, I am sceptical about the notion of a total spatio-temporal separation from the events and phenomena described in academic secondary sources. As Foucauldian genealogies of power and knowledge (Said, 1978; Mudimbe, 1988; Hall, 1992) demonstrate, the production of historical texts does not occur outside history, and it is in an important sense productive; writing history is not a simple and neutral revealing of the ‘facts’ of the past. As we shall see, this is a particularly pertinent observation in the cases of Burundi and Rwanda. As much is made clear by Chrétien (2003:14), who argues that the internalisation of the ‘colonial library’ and its use by ‘various groups who want to revive the previous century’s prejudices and orchestrate them politically’ – which in hermeneutic fashion is always an orchestration of the 101

past oriented towards the projection of the possibilities of the present into an indeterminate future – lies at the root of much of the violence in the postcolonial period.

Returning to historical events and processes via the secondary sources is not, then, equivalent to second-rate repetition. As Gadamer (2004 [1975]:xxxii) has shown with the idea of effective history, we confront this detail on a different plane of understanding and are moved by ‘different questions, prejudices, and interests’. Our approach to information is always filtered through our centredness and socialisation in particular spatio-temporal locations. We also ask certain questions of texts that constitute the basis of our approaching them. This need not be a hindrance. Without pre-interpretation or fore-understandings we would have no ‘entry- point’ into understanding; the information would be incomprehensible. As such, there is no access to things ‘in themselves’.

Of course, this presents methodological challenges. The work of historians, anthropologists and other area specialists is not somehow free of this. The ‘untapped sources’ of scholarship on non-Western world regions and civilisations (Arjomand, 2014:3), as presented in area studies literature, does not constitute a simple empirical store cupboard to be raided by social theorists, where things are revealed ‘as they essentially happened’, to paraphrase Leopold von Ranke’s dictum of nineteenth century historiography. Whether wittingly or unwittingly, such scholars approach their archival sources and subjects with particular questions and pre-understandings, undoubtedly informed by theoretical and normative categories (Hall and Tarrow, 2001:99).

For instance, it is interesting to note that area studies emerged principally in American university departments in the context of the Cold War where ‘the lack of global literacy became an issue of national concern in the harsh light of the incipient nuclear arms race and worldwide tension between the United States and the Soviet Union’ (Schäfer, 2014:148). From the early 102

1950s until the 1970s, US organisations such as the Ford Foundation funded an unprecedented amount of research in regions around the world48. There has been a profound change in area studies departments following the end of the Cold War; development programmes and global institutions favour the insights of economic and political specialists in the ‘global’ issues of the present (Greene, 2003:137; Schäfer, 2014).

This should not, however, imply that such work is fatally compromised. Indeed, the work of area specialists in painstakingly reconstructing the history under consideration is invaluable to my aims in this thesis. My approach is characterised by a conviction that this historical work – the result of careers spent in archives or on fieldwork assignments in places that the vast majority of the world cared little about until very recently – is the fruit of disciplinary, linguistic and methodological skill and dexterity in which it is worth investing trust. I have a deep admiration for this labour and have tried to do justice to it and the lives of those documented within it.

This does not preclude, however, a critical stance vis-à-vis this work. As Joseph Bryant

(2000:509) states, when working with secondary literature ‘a point should be made of reviewing the pertinent critical exchanges between historical or area specialists that arise over the uses and abuses of source materials, and of recording that information within their own synthesizing accounts’ (2000:509). The stakes of historical interpretation are extremely high in the contexts of Burundi and Rwanda, and debates within historical and area studies have a

48 This also gave it an affinity with modernisation theory (see discussion on pages 50-54). It should also be mentioned that many of the most influential examples of post-war historical-sociological scholarship outside modernisation theory come from scholars based in the US who were working in or in close proximity to these centres. We might list Charles Tilly, Theda Skocpol, Perry Anderson and Reinhard Bendix in addition to Immanuel Wallerstein, who began his career as an Africanist; Barrington Moore, whose research for his 1966 book, Social Origins of Dictatorship and Democracy, drew to a large extent on work that he undertook at Harvard University’s Russian Research Centre; and Benedict Anderson, whose work on ‘imagined communities’ was produced in Cornell University’s Southeast Asia Programme in the context of his research on Indonesia (Hall and Tarrow, 2001:99) 103

significance and resonance that extend well beyond the comfortable surroundings of universities. One need only read some of the barbed arguments over the years between the historian Jean-Pierre Chrétien and the political scientist René Lemarchand, two major representatives in the area studies literature, for an understanding of the sorts of debates that must be taken into consideration (e.g. Chrétien, 1990; Lemarchand, 1994:33).

In order to overcome these interpretive conflicts, I have referred to a wide range of secondary sources in order to see both points of wide agreement and points of disagreement.

Where agreement is present, all possible sources are referenced. Where exegesis is presented in relation to the material of a smaller amount of texts, often due to a lack of texts referring to a particular historical period or event, their quality and reliability has been ascertained and evaluated with the aid of book reviews or their prevalent presence in the literature elsewhere.

There are some scholars whose works are drawn upon across the entirety of the substantive empirical section of the thesis. These include such works as Réne Lemarchand’s Rwanda and

Burundi (1970) and Jean-Pierre Chrétien’s The Great Lakes of Africa: Two Thousand Years of

History (2003). These are seminal works of considerable depth and scope that treat long stretches of time, which explains the relative frequency of their appearance compared to others.

It should also be noted that some strands in the area studies literature specific to Burundi and Rwanda possess significance for revising and extending sociological concepts. Newbury and Newbury (2000:835), for instance, suggest that scholarship on the social history of central

Africa anticipates the subaltern school of postcolonial historiography in India and beyond, in its focus on decentring colonialist historians’ focus on states and elites by referring to ‘hidden’ histories, peasant relations and interpretive agency. Catherine Newbury’s (1988) work on the expansion of precolonial Rwanda presents some fascinating parallels with Elias’ theorisation of the processes of civilisation, the intertwined development of sociogenesis and psychogenesis, and the potential of this to be extended beyond European case studies (1994 [1939]). Elsewhere, 104

the challenges specifically raised in the literature of Burundi and Rwanda point to some interesting and productive substantive and epistemological challenges concerning the documentation of history and the integrative ambitions of social theory. In Rwanda, an historical oral literature accompanied the expansion of the Nyiginya kingdom, for reasons of both posterity and ideological legitimation. In Burundi, however, precolonial political development precluded the documentation of historical events in the same fashion. As the pioneering figure of oral history, Jan Vansina (1985:115; see also Lemarchand, 1970:304), suggests:

The political system [in Burundi] did not favour historical memory… It was in everyone’s interest to forget the past, whether it was the ganwa who had taken the land, the subchief who had been dismissed, or the king himself who relied now upon one faction, now upon another. The former senior regent of the country told me that history was of no interest at the court so there were practically no historical accounts. The political system shows why. The source materials used for empirically analysing historical societal traditions, for instance as in civilisational analysis, have political and social contexts that have to be hermeneutically reconstructed. As the case of Burundi demonstrates, the non-recording of history, written or otherwise, does not signal some pre-civilisational existence but instead reflects certain political features of societies and the role of historical documentation therein.

Overall, the hermeneutic interpretation of secondary texts can also work to bring together seemingly disparate areas of study. In Kinsella’s (2006) words: ‘The task is to find a common language through which the various texts can be given a voice to participate in conversation and speak to one another’. A legitimate hermeneutic task for any putative ‘global’ sociology, I contend, is that of bringing disparate bodies of work into dialogue or conversation with one another, not from the position of a ‘spectator’ (Dewey, 1929:23) but rather as a further participant in this dialogue. It is in this sense that my project is an essentially interdisciplinary activity, which involves bringing into dialogue literatures whose common separation can often 105

imply that they refer to different and incommensurable worlds rather than the entangled histories of a shared world.

LIMITATIONS, REFLEXIVITY AND ETHICS

There is a profound disconnect between writing about violence and colonial subjugation

– as documented in books, prefigured in historical sources or reflected upon in testimony – and having to live amidst the brutal fact of its reality. I have considered qualitative interviewing in this project, particularly in cases where there are ‘silences’ in the secondary literature and archive. I decided against it in this instance, although I am open to such work in a future research agenda49. This is largely due to the pragmatic difficulties entailed in doing research in each area considered and in the time-frame of the project. Another barrier to doing primary data collection is the ongoing regional instability, and the fact that attestations and research clearance must be sought in Burundi and Rwanda. These bring significant hurdles to bear on research, as some have documented (Thomson, Ansoms and Murison, 2013; Reyntjens, 2016), not least because the official versions of historical narration by post-genocide governments, especially Rwanda’s, makes it extremely difficult to access the ‘hidden transcript’, to use James

C. Scott’s term (1990).

I also have methodological reasons. This project concerns historical processes and constellationary events over a relatively long duration and as such necessarily eschews in-depth investigations of the short duration, though I am indebted to many exemplars of the latter.

There are rich layers of social reality that I do not claim to capture. To be clear, this by no means sidesteps the issue of research ethics. In its potential to cause complacency, in fact, it actually magnifies it. I have tried as far as possible to avoid objectifying the historical experiences under investigation. Where claims are made related to the motivations or

49 See pages 257-259. 106

normative orientations of individuals or groups, these are based on the careful consideration of the thorough research of others to whom I am indebted.

I have been troubled since starting this research by the notion that certain topics and events, including those under investigation here, might be ‘off-limits’ to theorising and ought perhaps to remain the reserve of specialists within African studies, of pragmatic and principled humanitarian and development institutions, or spoken only from the ‘standpoints’ of those who lived through them. However, I have increasingly come to believe that such thinking gives weight to an erroneous and deeply damaging idea that places like Burundi and Rwanda are of little importance or consequence to global history and modernity. Genocide is a crime against humanity, an affront to the universality of the rights and dignity of individuals, and colonial- imperialism is an offence against the rights of communities and societies to autonomous modernities. Each of these concerns the universal ‘we’ of humanity, and though we invoke this

‘we’ from particular locations that need to be accounted for to the extent that any claim to universality must be qualified and in principle open to dialogical questioning, we must nevertheless be able to invoke it (Harrington, 2000:35, 2016).

Of course, any study that addresses phenomena such as colonialism, mass violence and genocide, phenomena which stand as symbols for evil in various ways, must make its claims with humility and recognise that there are important elements that elude understanding

(Bernstein, 2002:7; Ray, 2011:171). But often the labelling of an event as ‘evil’ has been used as a way of avoiding the attempt to understand. The declaration of evil, of aberration, often functions as a way of shutting down in advance the disquieting idea that genocidal social practices might be more continuous with ‘everyday’ social practices; that they might be a latent possibility of the so-called ‘normal’ world rather than its absolute opposite (Bauman, 1989; see also Rose, 1996:43; Silverman, 2013:13). 107

So far, I have outlined the theoretical and methodological framework of this project. I have delineated what is meant precisely by key terms such as genocide and colonialism and have worked them into a theorisation of the multiplicity of modernity. It has been argued that

Burundi and Rwanda ought to be seen as specific and entangled routes to and through modernity. In this chapter, I have suggested that the tradition of philosophical hermeneutics constitutes a grounding in the philosophy of social science for contemporary programmatic endeavours to integrate social theory and area studies. The present thesis is an historical- sociological contribution to this task of integration. I have also detailed several methodological problems that present themselves specifically to this subject.

In the next four chapters, I will apply the theoretical and methodological framework outlined so far to distinct thematic and empirical objects. In chapter six, I look at the Rwandan and Burundian experiences of colonial modernity. In chapter seven, I analyse the multiple interpretations of the imaginary significations of modernity during the emergence of independence movements. Finally, in chapter eight, I consider the phenomenon of postcolonial violence in the light of the region’s specific route to and through modernity. Next, however, I consider precolonial social development until the eve of European penetration.

108

PART TWO

CHAPTER FIVE – PRECOLONIAL BURUNDI AND RWANDA: A HISTORICAL SURVEY

It has long been a trope of Eurocentric thought that African societies south of the Sahara possess no characteristics that could be said to be historical in any meaningful sense. For Hegel

(2001 [1837]:109-117), Africa was famously ‘the land of childhood, which lying beyond the day of self-conscious history, is enveloped in the dark mantle of night’; any world-historical civilisational legacies in the northern ‘coast-territory’ of Africa, ‘must be attached to Europe’, whereas the vast continent below ‘is the Unhistorical, Undeveloped Spirit, still involved in the conditions of mere nature… on the threshold of the World’s History’. This idea of teleological progress with the West as its torchbearer suffuses colonial texts. The British explorer of Central

Africa, David Livingstone (1865:6), wrote that ‘the first white man cannot but feel in a continent whose history is only just beginning’. Here, African precolonial societies were overwhelmingly represented as being cocooned in a more-or-less static and aeonian condition of tradition, and the ‘civilising missions’ of colonial-imperialism entailed their assimilation into world history.

The first European explorers to the Great Lakes region in the nineteenth century were taken aback by the development and complexity of the societies of Burundi, Rwanda and the region more broadly. As we shall see in the next chapter, they attributed this to a civilising invasion of ‘Hamitic’ peoples from the northern coast-region of Africa, who they named ‘Tutsi’.

From the point of encounter, the region became the subject of European written accounts that speculated on the emergence of kingdoms, the mixed economies of agriculturalism and pastoralism, and racial difference (D. Newbury, 2001:258). 109

In the 1950s, with the emergence of area studies departments and the extension of modernisation theory into the non-Western world, the Great Lakes region provoked a considerable amount of scholarship on its traditional societies 50 . As Chrétien (2003:26) suggests, the region offered ‘exceptional examples of centralized powers, social hierarchies, sacred politics, and ties of dependence. Its isolated position in the heart of the continent and its relatively recent contact with the external world… seemed to permit quasi-direct observation of an ancient way of being’. Of course, human settlements emerged way before their so-called

‘discovery’ in the colonial encounter, and this was recognised by early explorers. However, the explorers’ assumptions regarding the stasis and timelessness of these settlements prior to

European contact, as well as the thesis of racial ‘invasions’, have been thoroughly discredited.

They are now revealed as derived from a colonial logic that had a particularly strong role in the restructuring project of colonial modernity outlined in the next chapter.

On the contrary, it is important to stress that what have often been presented as static, timeless and undifferentiated societies (Reid, 2011) were in fact dynamic and transformational social formations. David Schoenbrun (1998), for instance, begins his longue durée analysis of the ancient socio-agricultural foundations of the Great Lakes region in 1000 BC, before identifying 800 to 1500 AD as a particularly noteworthy period of ‘the interplay of environmental, social, and philosophical change’, culminating in the emergence of ‘the Great

Lakes as a Bantu-speaking region’ (Schoenbrun, 1998:262; see also Chrétien, 2003:58). The kingdoms at the roots of present-day Burundi and Rwanda have been traced using oral sources to the period around 1350-1500, developing alongside a variety of other societal forms ranging from tributary societies to empires (Bishikwabo, 1991:15; Mworoha and Mukuri, 2004:71;

Hintjens, 2008:1669). By the nineteenth century, Burundi and Rwanda were complex

‘monarchical states’, each with around two million inhabitants (Schoenbrun, 1998:258;

50 See discussions on pages 51-52, 104. 110

Chrétien, 2003:158). This precolonial societal formation cannot be seen as a teleological progression towards the state of affairs on the eve of European encroachment (Chrétien,

2003:43). The colonial entanglement began at a point when the polities in the area were in a period of crisis and transition. These crises were in part the result of regional political struggles, but there were also important environmental and ecological factors (Botte, 1985, 1985a;

Thibon, 2004:ch.2).

In this chapter, I will present the precolonial historical development of Burundi and

Rwanda in comparative perspective. I will delineate their differences and similarities in terms of underlying cultural patterns and the social and political development of the monarchical states of the nineteenth century. I will begin by situating them in the context of the Great Lakes region, before discussing Rwanda and Burundi in turn and finally culminating the discussion in a brief comparative section.

THE GREAT LAKES REGION

Burundi and Rwanda are located in the Great Lakes region of Central Africa, an amorphous area in both an epistemological and geographical sense. The term, related to the earlier expression of ‘interlactustrine area’, comes from the German noun Zwischenseengebiet

– ‘the territory between the lakes’ – which was fantasised about and charted by European explorers. When searching for the source of the Nile, they ‘discovered’ large mountainous areas and lakes in inland Africa that correlated with Alexandrian maps. These maps were themselves rooted in the ancient Greek quest for the discovery of the sources of the Nile infamously captured in Ptolemy’s speculative descriptions of the ‘Mountains of the Moon’, widely thought to be the Rwenzori Mountains on the border of western Uganda and eastern Congo (Chrétien,

2003:22-23, 2010). The ‘Great Lakes’ is thus a term implicated in colonialist fantasies and in 111

what Valentin Mudimbe (1988) called the ‘invention of Africa’ (see also Omobowale,

2015:108).

Over time, however, the moniker has come to be deployed by European and African scholars and in the titles of local, regional and global institutions, to refer to a cluster of central

African nation-states. Within academic area studies, there is widespread agreement that this region includes, in addition to Burundi and Rwanda at its centre, eastern Congo, south-western

Uganda and north-western Tanzania (Chrétien, 2003; Lemarchand, 2009:3). A wider definition is provided in the United Nations’ (2013) Peace, Security, and Cooperation Framework for the purposes of bringing peace to the Democratic Republic of Congo and the wider region. It was signed in 2013 by Angola, Burundi, Central African Republic, Congo, the Republic of

Congo, Rwanda, South Africa, South Sudan, Tanzania and Uganda. However, the Economic

Community of the Great Lakes Countries, established in 1976 as a regional framework for economic integration, includes only Burundi, Rwanda and the entirety of DR Congo, known at the time of its formation as Zaire (United Nations, 1978). In addition to its shifting geographical boundaries, the cultural unity connoted by the term ‘Great Lakes region’ is further compounded by ‘the diversity of peoples, cultures, and sub-regions subsumed under this label’ (Lemarchand,

2009:3).

Nevertheless, there is perhaps a case to be made for the Great Lakes region’s ‘regional’ or even civilisational status, when defined as ‘enduring ancient, large-scale cultural formations surpassing individual polities’ (Therborn, 2014:267) comprised of a common cosmology, a set of symbolic, linguistic, mythical and aesthetic factors, and trans-ethnic affinities of social and economic practices. As Therborn (2011:7) has noted, following Braudel and evoking the

‘regional sociology’ of Radhakamal Mukerjee (1926), civilisations or regions are spatial entities. The limiting and enabling conditions of climate, ecology and topography are important factors in the emergence of kingdoms and the distinct forms of social relations in the region. 112

For instance, due to their mountainous topography, the colline (or hill) has historically been the central locational unit of social life in both Burundi and Rwanda (Mworoha, 1987:38-39;

Uvin, 1997). Furthermore, time was codified and recorded in agricultural calendars structured around planting and harvesting festivals. The most prominent of these was the festival concerning sorghum, in Burundi called umuganuro and in Rwanda called umuganura. As

Schoenbrun (1998:254-255) states:

In their ceremonial structure and personnel, these festivals performed prevailing moral sentiment and social hierarchy. More importantly, they were chances for people to struggle over the content of those two conditions. The planting and harvest festivals of the uplands were also opportunities for negotiating or renegotiating the social basis of political and ritual power. Related to regional technological changes in agriculture, cattle-raising and metallurgy was the emergence of a linguistic region comprised of variations of Bantu languages (Schoenbrun,

1998; Chrétien, 2003:45; Therborn, 2011:24). Burundians speak Kirundi and Rwandans speak

Kinyarwanda; they are distinct languages, but mutually intelligible.

Though nineteenth century philologists like Wilhelm Bleek suggested that the Bantu language was unable to aspire to poetry or to philosophy (Chrétien, 2003:48), distinct ‘cultural programmes’ (Eisenstadt, 2000) emerged in this context. The idea of a ‘Bantu philosophy’ is highly contested (Mudimbe, 1988:135-187; Pearce, 1989; Therborn, 2011:26). It was first named as such in the title of a book by a Belgian missionary, Father Placide Tempels (1959).

Its essential argument was that, in contrast to racial anthropological conceptions of the

‘primitive mind’, African social organisation could be imputed to a comprehensive Bantu

‘philosophy of life’ or ‘ethnophilosophy’, which revealed itself in language. Bantu ontology was based on the premise that the universe is constituted by dynamic and hierarchical ‘vital forces’, as distinct from the Western ontology of ‘being’:

When we think in terms of the concept of ‘being’ they use the concept ‘force’. Where we see concrete beings, they see concrete forces. When we say that ‘beings’ are differentiated by their essence or nature, Bantu say that ‘forces’ differ in their essence or nature. They hold that there 113

is the divine force, celestial or terrestrial forces, human forces, animal forces, or mineral forces (Tempels, 1959:52). The idea of a singular Bantu ethnic group has been discredited and the thesis is heavily imprinted by what Gadamer (2004 [1975]:275) called ‘romantic revaluation’51. Nevertheless, these statements correspond to ‘certain widespread themes of beliefs of contemporary significance [in sub-Saharan Africa]. There is a rich, diversified spiritual world, of both malevolent and benevolent spirits, to which humans owe respect and sacrifice. These are the spirits of ancestors, and there are spirits of nature’ (Therborn, 2011:26). The value of fertility in African societies is well-established (see discussion by Therborn, 2011:33), and in precolonial Burundi and Rwanda, these ‘abstract concepts of fertility and fecundity were embodied in cattle, food, hoes, and children’ (Schoenbrun, 1998:257). There are also striking continuities with the work of the anthropologist Christopher Taylor (1988, 2004) on the ontological concept of ‘flow’ in precolonial Rwanda, and the negativities associated with blockage or solidification52. This takes on a high level of significance when we consider, as I argue in chapter six, that the racialisation of tradition which occurred under indirect rule in

Ruanda-Urundi had a ‘solidifying’ effect on social relations.

Precolonial Burundi and Rwanda also shared patterns of cultural expression, including similar traditions of drumming, singing and dancing centred in the royal courts. Genres of oral expression – riddles; proverbs; plays; chanting – that narrated the past became sites for the

51 I employ this to refer to Tempel’s contradictory assumption that while Bantu and Western traditions and social formations were formed against the background of completely different ontologies, he himself was somehow able to suspend his formative cognitive structuring in the Western tradition in order to inhabit and translate the Bantu tradition. 52 In his studies of popular medicine in Rwanda during the 1980s, Taylor (1988) observed that illnesses were often characterised by perceived irregularities in flows of fluids, and impotent men and infertile women were stigmatised for representing ‘blockage’. The flow/blockage metaphors, Taylor argued, were also present in the rituals of kingship and the idea of the mwami (king) as a catalyst of flows and eliminator of blockages (women of child-rearing age who had not menstruated could be killed). Ideas of flow and blockage pervaded the cosmological order represented by Imana. Imana, often translated as God (shared by both Burundians and Rwandans), is more accurately translated as a transformative or creative force (Taylor, 1999:113-126). 114

legitimation of royal authority and ‘with the establishment of politically centralized kingdoms, the hierarchy of kings, chiefs, and priests moved to establish authorial control over particular sorts of content for these general types of speech’ (Schoenbrun, 1998:259). In Rwanda, the narrators of history at the court were known as the abiru53. They were interpreters of the rituals and codes (ubwiru) of the monarchy and enacted history through ‘telling’ in the form of theatre

(ikinimicu). In doing so, they played a primary role in judicial interpretation, and ultimately in promulgating a social order based on the centrality of the kingdom and on the mwami54 as a conduit for the vital force of imana (Breed, 2006:508). In Burundi, the interpreters of the historical and judicial underpinnings of the court were represented by the banyamabanga

(custodians of court rituals) and the abashingatahe (elderly guardians of the legal system)

(Ntahombaye and Nduwayo, 2007:244). These forms of knowledge worked in tension with those propagated by actors distinct from the courts, such as spiritual healers, and those emanating from societal formations in the kingdoms’ peripheries (Schoenbrun, 1998:259).

Overall, as J. P. Chrétien (2003:41-42) argues:

On the various canvases of so-called traditional African civilisations, the Great Lakes region seems to stand out for its high degree of cohesion. The region has centralised polities, common religious references, intense human density, and similar languages, all of which underlie a contrast with neighbouring areas. These traits suggested a distinct cultural personality, at least in the state in which they were directly observed during the nineteenth century. One should be careful, however, of containing history at the regional level. These societies were certainly embedded within the entangled histories of the wider Great Lakes region and connected by trade and commercial networks (D. Newbury, 1980; Wagner, 1993; Chrétien,

2003:191), ecology (Botte, 1985, 1985a), migration (Webster, 1979) and conflicts (Chrétien,

2003:160). At the same time, there were significant levels of variation among Burundian and

53 Alexis Kagame (1912-1981) was one of the last and most prominent of the court historians and it is due to his work that so many of Rwanda’s oral traditions are written (see Kagame, 1951). 54 Mwami is the word equivalent to ‘king’ in the languages of Burundi and Rwanda, Kirundi and Kinyarwanda respectively. 115

Rwandan precolonial societies concerning political centralisation, modes of social organisation and the nature of social identities therein (Lemarchand, 1970; Mamdani, 2001:56; D. Newbury,

2001; Daley, 2003:45). In the remainder of this chapter, I focus on the political development of the monarchical states of Rwanda and Burundi, and their intertwinement with the development of social identities. I make comparisons between the trajectories of the two countries and highlight some significant factors for the themes taken up in the following chapters.

RWANDA

As a political entity, the Rwandan state can roughly be traced to the middle of the seventeenth century (C. Newbury, 1983:258; Vansina, 2004:207). Over the course of the eighteenth and nineteenth centuries, it grew from a small polity into a sizeable monarchical state (Chrétien, 2003:139) more-or-less coterminous with its present-day borders. This was not an inevitable, wholly intentional and unidirectional historical movement. As Catherine

Newbury (1983:257-258) states, ‘the amount of territory controlled by the kingdom fluctuated as a result of succession struggles, conquest by enemy powers, and at times successful annexation’. However, she continues, ‘beginning in the eighteenth century Rwanda experienced a steady expansion in both territorial scope and royal power’ (see also Chrétien,

2003:158-159 and Vansina, 2004:188-189).

At the time of European exploration, Rwanda exhibited a high degree of political centralisation and hierarchical organisation. The population – though structured into lineage, clans and ethnic groups and regionally differentiated – shared the same language and culture

(Twagiramutara, 1998:107-108). The roots of these features can be traced alongside the history of Nyiginya kingdom. Nyiginya is one of the clans (ubwoko) of Rwanda. It was one among a variety of historical polities of distinct sizes and characteristics, ‘from extensive kingdoms of 116

the east (such as Gisaka, Ndorwa, Bugesera, Mutara) to the smaller autonomous political organizations of the west’ (Newbury and Newbury, 2000:850). The kingdom emerged as the forerunner on account of its organisation and structural innovation, and played a prominent role in shaping Rwanda as confronted by European explorers in the late nineteenth century.

Particularly pertinent is the period between the reigns of mwami Rujugira (c. 1770 -

1786) and mwami Rwabugiri (c. 1860-1895). The time that elapsed between these reigns saw the intertwined development of processes of military expansion and political centralisation.

Military development in this period can partly be attributed to a response to the combined threat of Rwanda’s neighbours, in addition to the conquest and subjugation of peripheral kingdoms and their absorption into the core; political centralisation was something of an unplanned process, contingent on warfare for both internal structural changes and for projecting an image of itself externally (Prunier, 1995:15-18; Mamdani, 2001:67). Lemarchand (1977:71-72) defines the development of the precolonial Rwandan state out of the Nyiginya kingdom an

‘indigenous form of imperialism’ (see also Newbury, 1988; Des Forges, 1995).

As the influence of the kingdom grew militarily and politically, there emerged a definite elaboration of a court sensibility based on aesthetic and linguistic patterns of differentiation from the kingdom’s subjects, consciously distinct from the ‘commoner’ culture of the periphery

(D. Newbury, 2001:293; see also Chrétien, 2003:161; Vansina, 2004:180, 197). This was institutionalised in the codification of royal rituals (ubwiru) and in the role of the abiru who expounded an historical oral literature replete with origin myths and historical poems

(bitekerezo) (Chrétien, 2003:161; Kamitali, 2014:157). The most famous of these is the myth of Gihanga, the legendary first king of Rwanda. Gihanga came down from the heavens and had three sons named Gatutsi, Gahutu and Gatwa (Tutsi, Hutu and Twa). He entrusted each son with a full pot of milk and asked them to guard it until his return. Gatwa drank the milk and fell asleep, and Gahutu spilt some of his by accident after falling asleep. Only Gatutsi exercised 117

the requisite vigilance to keep his pot full, and he was therefore granted the right to rule Rwanda

(Stewart and Strathern, 2002:29-30; Gatwa, 2005:51-52)

The expansion of the court was also accompanied by the migration of people in search of new lands from the centre to the periphery, a process intertwined with the development of land tenure systems and centralised tributary taxation (Chrétien, 2003:160-161, 183). These settlers acted as informal agents of the kingdom. As Chrétien (2003:161) states, ‘in these distant lands they commended themselves to the central authority in such a way that it could be useful to them, while, in return, they became agents of that authority’s influence’.Through this mechanism, he continues, ‘the image of the centre was constructed in the border zones, because the peripheral populations who had lived there longer were perceived pejoratively as not very

‘Rwandan’. They served as a foil to the image of central civilization, which was defined in counterpoint’ (see also C. Newbury, 1988; Newbury and Newbury, 2000).

It is in this context of core-periphery relations and processes of state formation that the development of the social identities of Hutu, Tutsi and Twa55 must be situated. Far from being fixed, primordial groups that had migrated to or invaded the region at a distinct point in time – which, as we shall see, was a core assumption of the Hamitic Hypothesis formulated by

European explorers, though its influence far transcends this context56 – these were identities that were produced and reproduced in institutionalised practices, and whose usages changed

55 I focus here in particular on the social identities of Hutu and Tutsi, because they are most pertinent for the overall themes of the thesis. However, I must at this point acknowledge the Twa, who have long been the lowest-ranking social group in both Burundi and Rwanda (and, comprising a mere 1% of the population in each, the least numerous social group). In Rwanda, the persecution and marginalisation of Twa extends back into the precolonial period (Taylor, 2004, 2011; Ndahinda, 2011:215-250). 56 (See, for example, Abrams, 1995:144-145; Magnarella, 2005:802-803). There are no doubt long histories of population movement in the region, caused among other things by climatological shifts, famines and droughts. However, the positing of these movements as dramatic and abrupt invasions of ‘Bantu agriculturalists’ and, later, ‘Nilo-Hamitic pastoralists’ (i.e. Hutu and Tutsi respectively) is erroneous and, as recent history attests, dangerous in its ideological implications. It is much more appropriate to talk of ‘micro-migrations’ spread over time. The ‘ethnogenesis’ of the categories of Hutu and Tutsi cannot be divorced from their complex and in many ways contingent historical development in line with processes of state formation (Chrétien, 2003:41-83). 118

over time in line with the processes of political centralisation and military expansion outlined above.

They were also, however, identities linked to the culture of the court. As implied in the origin myth of Gihanga, the term ‘Tutsi’ came to denote a certain relationship to power. It emerged initially as a self-referential term used by a fraction of pastoralists before coming at the time of European arrival to refer to the political elite of the Nyiginya kingdom. The term

‘Hutu’ also developed as ‘a demeaning term that alluded to rural boorishness or loutish behaviour used by the elite’ and, further still, was applied without discrimination to all those outside of the realm of the kingdom (Vansina, 2004:134-135; see also C. Newbury, 1988). In this sense, the central model was culturally defined in large part by its relationship with otherness in the peripheries. One has to note here the similarities between the

‘civilised/barbarian’ scission that so heavily imprints such relationships as those between the

European metropole and the colonial periphery. Civilising processes and the intertwining of state formation and habitus are not exclusively European or Western phenomena (Elias, 1994

[1939]; Kilminster, 2007:70; see also the example of Chinese high culture and ‘barbarians’ in

Osterhammel, 2014:827).

The usage of the terms ‘Hutu’ and ‘Tutsi’ in this sense has been dated to the 16th century

(Nkurikiyimfura, 1994; Vansina, 2004:271), and developed alongside the expansion of the

Nyiginya kingdom in the relational sense outlined above. However, the first institutionalisation of these social categories occurred under Rujugiri’s rule in the eighteenth century, when the terms came to refer to the military distinction between combatants (Tutsi) and non-combatants

(Hutu) (Vansina, 2004:135). In the nineteenth century, the shift in their meaning was embodied by another institutional differentiation. Under the administrative headings of ‘chiefs of the long grass’ (umunyankenke) and ‘chiefs of the land’ (umunyabutaka), there emerged a labour distinction corresponding to a differentiation between pastoralists (Tutsi) and agriculturalists 119

(Hutu) (C. Newbury, 1980; Vansina, 2004:132; Ngabirano and Tsimbi, 2012:50-51; Kamatali,

2013:159-160).

This distinction took on an increased significance when mwami Rwabugiri introduced uburetwa, a corvée system overseen by Tutsi land chiefs with links to the royal court. This was extended outwards from the centre of the kingdom and eroded the autonomy of ‘Hutu’ lineages with tenure over the land; tenure henceforth became conditional upon payment of prestations to the court in the form of food and labour (C. Newbury, 1980:100). During Rwabugiri’s reign, it gradually expanded from one day out of six to three, and it undermined the livelihood and security of the peasantry. It had an especially profound effect on Hutu-Tutsi relations, because it overwhelmingly affected poor Hutu people, who comprised the bulk of the population

(Pottier, 2002:13). This land tenure system was humiliating and constituted the apogee of the

‘scission of society into the Tutsi and Hutu social categories as a case of disagreement between a ruling class and its subjects’ (Vansina, 2004:136, 191-192; see also Rwabukumba and

Mundandagizi, 1974; Vidal, 1974).

It was frequently argued by scholars of late colonial and early independence-era

Rwanda that the Hutu-Tutsi schism was suffused through society via a mechanism called ubuhake (e.g. Maquet, 1961; for discussion see Chrétien, 2003:187). This was a social institution represented by the granting of cattle from a patron to a client, a relationship which supplemented the categorical divide that from the eighteenth century increasingly came to distinguish pastoralists and agriculturalists, and which provided ideological justification for a

‘premise of inequality’ between them. It has since been convincingly demonstrated, however, that ubuhake was highly localised in the areas around the central court and far less frequent than earlier scholars had assumed (e.g. .C. Newbury, 1983). 120

Nevertheless, it played a distinct role in the colonial administration of the country, and of colonial-era scholarship. In the notion of the ‘premise of inequality’ that was influential in

Rwandan studies for a very long time, ubuhake came to be seen as a sort of cultural coagulant that provided cohesion to, in David Newbury’s (2011:xxvi-xxvii) words:

a disparate society formed of groups that were perceived as distinct in racial, historical, and cultural terms… Incorporating ethnic hierarchy, occupational diversity, and political centralization into a clear and comprehensive model, this vision formed a perfect static image – one that met the goals of the colonial administration as well as those of the Court and justified the parameters of colonial rule (see also Nelson, 1983). As we shall see, these social identities and the tension between them were racialised in the ideas bound up with the Hamitic hypothesis. The racialized identities were in turn institutionalised during the rationalisation of Ruanda-Urundi that occurred under Belgian colonial rule, particularly in the administrative reforms from 1926 onwards. Both the institutions of uburetwa and ubuhake were extended and racialised in these reforms. The extension of the latter is particularly important. To quote at length from Newbury and Newbury

(2000:861):

Ubuhake clientship was infrequent in the precolonial period, not widespread – and certainly not universal – as the conventional understanding asserted. For example, at the end of the nineteenth century, at the height of precolonial state institutions, only 8 percent of family heads were involved in ubuhake ties at any time in their lives, according to a careful study carried out in south-central Rwanda, near the cultural heartland of the kingdom. Far from being universal, ubuhake was of very limited extent. Furthermore, according to the precepts of ‘modernization theory’, during colonial rule, as individuals are released from the responsibilities placed on them by archaic institutions, ‘feudal’ institutions such as clientship would be expected to diminish in their extent. Yet the data suggest something else: over the first generation of colonial rule, ubuhake ties in this sample actually increased, almost doubling the percentage of family heads involved; therefore clientship appears to have been extended and reinforced by colonial power. This is significant in terms of the content of the next chapter, in the sense that it constitutes some empirical proof of how the assumptions made in colonial discourses about tradition led to its material consolidation and extension. For now, it is worth saying that, though increasingly rigidified during the nineteenth century under Rwabugiri’s reign, Hutu and Tutsi were to an extent ‘fluid’ identities prior to colonialism, as demonstrated by the precolonial practices of 121

kwihutura, where Hutu could become Tutsi, and gucupira, where Tutsi could become Hutu, ordinarily by wealth or status being accrued or lost,, respectively (Maquet, 1954:120; Mamdani,

2001:70). However, it is simply wrong to say that Europeans invented these categories or that interethnic violence was set in motion solely by the policies of colonial powers (e.g. Gourevitch,

1998:59).

The Hutu-Tutsi distinction, then, sharpened over time and manifested in outright localised social conflict under Rwabugiri’s rule, with various incidences of insurrection and counter-insurrection occurring in the 1880s and 1890s. Crucially, some of these (notably an armed movement which broke out in north-western Rwanda in 1897) were explicitly anti-Tutsi

(Vansina, 2004:136-139). It is important to point out that this precolonial social conflict was not conceived in racial terms; rather, it was a response of a subjugated group to the oppressive incursion of the central court into an under-integrated region. To say so is not to sanitise the past and absolve colonial powers of any responsibility for postcolonial violence (cf. Daley,

2006:663). Instead, it is to point out that the racialisation of the Hutu-Tutsi schism solidified an existing state of social conflict which had crystallised around monarchical power. As Jan

Vansina puts it: ‘the overwhelming impression one is left with after a study of the age of

Rwabugiri and his successors is that of the unrelenting rise of a tide of terror that starts at court and engulfs the whole country, finally erupting into a social crisis that has now lasted for well over a century’ (Vansina, 2004:164).

Overall, Rwabugiri’s reign gave rise to structural transformations within Rwanda, which were vital in establishing the static image of Rwandan society at the time of European arrival (D. Newbury, 2001:307; Vansina, 2004:126). The political borders of precolonial

Rwanda, and the contours of its social divisions, were thus by and large in place before

European colonisation, barring some autonomous communities in the North and West, which were annexed in the twentieth century with the aid of the colonial powers of Germany and then 122

Belgium (Lemarchand, 1970:22). In these relatively autonomous parts of present-day Rwanda at the turn of the twentieth century, there was no broad social category uniting Hutu (C.

Newbury, 1988; D. Newbury, 2001:273).

It was in the context of incorporation into the kingdom in the nineteenth century, aided significantly by the German administration in the early twentieth century and later bolstered by Belgian colonial reforms replete with censuses and the statistical measurement of the population, that parts of the population, hitherto defined along regional or clan lineages, became Hutu. However, even when integrated into central Rwanda, these communities, in particular the north-western Kiga, retained a high degree of cultural autonomy and animosity to the Nyiginya kingdom, and towards the end of the nineteenth century a number of messianic resistance movements such as the Nybingi cult, which would later take on an anti-colonial imperative, arose within them (Ranger, 1968:451; Chretien, 1972; Freedman, 1974;

Lemarchand, 1977:73-74; Rutanga, 1991). These regional identities retained a strong significance in independence movements and in the post-independence government of the

Second Republic established in 1973; Hutu of Northern Kiga lineages formed the basis of the akazu, or ‘little house’, responsible for the architecture of the Rwandan genocide of 1994 (Des

Forges, 1999:39).

Rwabugiri died in 1895, whilst returning from an expedition to Bushi, a kingdom in today’s South Kivu province in eastern Congo, bringing to an end an era of economic and military expansion. His death was followed by a period of interregnum in the country and political turmoil in the central court as the rival dynasties fought for power (D. Newbury,

2001:310-311; Chrétien, 2003:173), culminating in the ascension of the young Yuhi Musinga

(Des Forges, 2011). It also meant the immediate reconstitution of areas of present-day eastern

Congo such as Bushi and Ijwi Island, into which Rwandan armies had moved during the 123

nineteenth century57. As detailed in the oral literature by one storyteller: ‘All the foreign countries that Rwabugiri had defeated reconstituted themselves, all of them! No foreigner remained within the country. They all reconstituted themselves right away’ (in Vansina,

2004:177). This coincided with the arrival of European powers and increasing outside incursions, notably involving the Force Publique, the army of King Leopold II’s Congo Free

State. Though the Rwandan military tried to halt these incursions, they were ultimately overpowered and lost some of their most celebrated military figures. The Rwandan court’s hitherto cautious, isolationist and at times confrontational stance towards outsiders was abandoned in favour of a model of co-operation with the colonisers (D. Newbury, 2001:311; see also Vansina, 2004:174-178).

It is in this context of crisis that the colonial entanglement in Rwanda must be situated.

Jan Vansina (2004:197), the foremost historian and anthropologist of central Africa and pioneer of oral historical methodology, strikes a blow to the claim that Rwanda was a united and harmonious country prior to the ‘divide and rule’ policies of the European colonisers58:

Far from constituting an apotheosis of a great united nation encompassing almost two million people, the kingdom of Rwabugiri and his successors offered the spectacle of nearly two million people standing on the verge of an abyss. The successive colonial governments racialised this social conflict and, in a fateful underappreciation of its complex historicity, assumed it to be a timeless feature of Rwandan social and political life.

57 The legacy of these imperial-military excursions can be seen in declarations by RPF government officials in the midst of the Congo wars which began in 1996. After the 1994 genocide, many Hutu civilians sought refuge in the neighbouring Congolese Kivu provinces. They were accompanied by many of those responsible for perpetrating the genocide, who used the refugee camps as a means of controlling the Hutu population and of launching murderous raids in western Rwanda. In 1996, the Rwandan government moved into the refugee camps and violently cleared them, pushing the militants further into the DRC and killing many thousands of those caught inbetween, raising questions of genocide (UN High Commissioner for Human Rights, 2010; Lemarchand and Reyntjens, 2011; for an indepth study of the Congo wars and the catalysing effect of the Rwandan genocide, see Prunier, 2009). This action was justified by some with irredentist arguments about the territorial heritage of the Nyiginya kingdom and its expansion into eastern Congo (D. Newbury, 1997) 58 This is the view presented in the exhibitions at the Kigali Genocide Memorial Centre. 124

BURUNDI

Burundi, like Rwanda, represents something of an anomaly in Africa inasmuch as it too was not a colonial construct. Like its northern neighbour, its boundaries were slightly extended and fortified under colonial rule but, by and large, its territory was in place before the establishment of the colonial state, while its people shared the same language and culture and were subjects to a mwami (Rutake and Gahama, 1998:80-82; Daley, 2006:662). Burundi also approached the end of the nineteenth century marked by political crisis and conflict. However,

Burundi’s character was sharply distinct from that of Rwanda. The centralised and hierarchical pattern of authority that characterised Rwanda at the end of the nineteenth century was not replicated here. What is more, there was a clear contrast in the roles of the monarchy in the social imaginaries of the two countries. In Rwanda, the expansionist ambitions of the monarchical state were a source of social conflict, whereas in Burundi the institution had a more unifying, symbolic role. Describing the ‘ethos of mwamiship’, Lemarchand (1970:303) states that:

Although the incumbent may have been regarded by some as unworthy of holding office, the office itself conveyed an image of virtue which clearly transcended the realm of human contingencies. It was part of a psychic whole which gave meaning and purpose to all members of society, regardless of social or ethnic origins. It was linked to the well-being of the nation through specific rituals (like those attending the celebration of the umuganuro, the sorghum festival) and symbols (like the royal drum, Karyenda), and thus acted as a mystic bond, unifying the dead, the living and the unborn, and, likewise, gave a sense of corporate unity to Tutsi, Hutu, and Twa. Relatedly, precolonial social identities, though containing some similarities, for instance in the existence of Tutsi, Hutu and Twa identities, were also decidedly different

(Prunier, 1995a:128) in Burundi and Rwanda. There are 220 clans in Burundi that run across the social identities of Tutsi, Hutu and Twa; these preceded kingship and continue to be influential today (Rutake and Gahama, 1998:81; Nsanze, 2001:4). Additionally, and to a greater degree than in Rwanda, there were distinct and important social and regional variations among Tutsi and Hutu (D. Newbury, 2001:274). For instance, within the Tutsi category in 125

precolonial Burundi, there were at least three layers, each intersecting with regional and class stratifications: Tutsi-Banyabururi, from the southern province of Bururi; Tutsi-Banyaruguru, dispersed throughout the country, but of higher social standing due to their links to the monarchy in Muramvya province; and Tutsi-Hima, a pastoral group of lower social rank

(Mafeje, 1991:58; Rutake and Gahama, 1998:80; Daley, 2006:662). Hutu people were also stratified internally, generally along occupational lines. At the threshold of the twentieth century, Burundi’s Hutu people were far more involved in the workings of government than in

Rwanda, in roles like the abashingantahe adjudicators or banyamabanga court ritualists.

The most important political distinction from precolonial Rwanda is the existence of an aristocratic group known as the ganwa, traceable to the emergence of kingship and comprised of four different dynastic lineages: the Batare, Bezi, Bataga and Bambutsa, associated with the names Ntare, Mwezi, Mutaga and Mwambutsa respectively (Lemarchand, 1966:405; Rutake and Gahama, 1998:103; Ndarishikanye, 1999:233). The Burundi kingdom was established under the reign of mwami Ntare Rutatshi, around the end of the seventeenth century (Chrétien and Bahenduzi, 1990; Rutake and Gahama, 1998:81-83). It was under his rule that a monarchical institution emerged, comprised of the tripartite consolidation of military organisation, religious worship and clanic solidarity, and began to expand and combine hitherto distinct areas into a shared historical field (Mworoha, 1987:120; D. Newbury, 2001:282).

As in Rwanda, the monarchical state was consolidated in the 18th and 19th centuries

(Chrétien, 2003:141), in particular under the reign of Ntare Rugamba (1796-1850). His rule

‘set the geographical contours to modern Burundi by extending royal power in many directions’

(D. Newbury, 2001:284; see also Nsanze, 2001:5). Ordained by the mwami, who was more a sacred figure – a ‘father of the nation’ (Prunier, 1995a:128) – than a political ruler, the ganwa formed a political-aristocratic elite and governed the large regions and populations of these new areas of the kingdom (Abrams, 1995:145; Rutake and Gahama, 1998:102; Chrétien, 126

2003:163). Power was thus decentralised and fragmented among relatively autonomous political units, each under the authority of a ganwa who sat atop a pyramidal structure of chiefs and sub-chiefs, and whose claims to authority ultimately depended on the genealogical remoteness or proximity of the dynasty from which they descended (Lemarchand, 1970:22,

1977a:96; see also Trouwborst, 2008:166).

Political conflict in precolonial Burundi tended to be framed by struggles between and within these princely factions (Lemarchand, 1994:36-39). This conflict became increasingly salient in the nineteenth century. With the death of mwami Ntare Rugamba in the early 1850s, and in the ensuing power vacuum, political culture became marked by fragmentation. Regional autonomy was very pronounced (D. Newbury, 2001:281) and conflict emerged between descendent lineages tied to Ntare, known collectively as the Batare, and those linked to Mwezi, known as the Bezi. Control over the territories expanded into was assumed by Ntare’s eldest sons – the Batare – and his youngest son, mwami Mwezi Gisabo, assumed power in Ntare

Rugamba’s former court (D. Newbury, 2001:284).

Just prior to the colonial encounter, then, Burundi was split into distinct spheres of political influence. The eastern region, approximately one third of the country, was under the sway of Batare princes. The mwami’s influence was limited to the core area of kingdom at the centre. Forming a buffer zone between this core area and the Batare-controlled east were the domains of Mwezi’s sons, the Bezi princes. Finally, covering an elongated strip of territory running on a north-south axis along Lake Tanganyika and the Ruzizi River were various semi- independent chiefdoms under Hutu or Tutsi control (Lemarchand, 1994:37; D. Newbury,

2001:285). 127

Burundi on the eve of colonial penetration was thus also a society marked by internal conflict and power struggles, principally between the Batare and Bezi ganwa factions. These political conflicts, according to David Newbury (2001:285):

represented not only conflict among Baganwa personalities, but true popular movements asserting limits to court penetration; they represented complex combinations of individual ambition on the part of the Ganwa, with a sense of class awareness and regional autonomy on the part of the cultivators. As we will see in chapter seven, this conflict between aristocratic factions burst forth in the country’s trajectory towards independence, infused with the imaginary significations of autonomy, social justice, national liberation and so on.

Whereas in Rwanda, as we have seen, the political conflicts at the central court suffused more generally into social relations in the nineteenth century during Rwabugiri’s reign, this did not materialise into polarised social conflict in Burundi. Unrest tended to be oriented towards the ganwa chiefs and the political structure around them. There was also fundamental ambivalence concerning the nature of Tutsi and Hutu identity in Burundi’s precolonial society, which militated against collective mobilisation in their name (Laely, 1997:699). In the Kirundi language, the term Hutu has two separate meanings, one referring to its cultural underpinnings, the other to its social connotations. In the social sense, Hutu refers to a subordinate in relation to somebody of higher standing. Thus a Tutsi who was a client of a wealthier patron would be referred to as Hutu, even while his cultural identity remained Tutsi (Lemarchand, 1994:11-14).

In the postcolonial period, these two meanings came to be virtually synonymous, but before colonialism there were none of the institutionalisations of this social differentiation seen in various forms in Rwanda (Chrétien, 2003:190). Status differences also occurred within clan lineages and affiliations, again stretching across the simplistic Hutu-Tutsi dichotomy

(Mworoha, 1987:175-178). 128

Patron (shebuja) and client (mugererwa) ties possessed a great significance in precolonial Burundi, in the form of the mechanism of ubugabire. Similar to the ubuhake clientship in Rwanda, ubugabire was a contract by which a donor made a gift of cattle to a beneficiary, who in turn undertook to fulfill certain obligations, ranging from bringing beer or food to political support or military service (Laely, 1997; Gahama, 2001:321). To be a client dependent on the benevolence of a patron was not necessarily negative. On the contrary, this dependence was a vector of social mobility in the acquisition of status to which, as we have seen, the designation of Hutu or Tutsi was attached. As Thomas Laely (1997:704) writes:

The structures of government and administration of the precolonial monarchy were not determined by permanently given territorial units; they were dependent on personal relationships of the moment, and were shaped according to the pattern of patron-client dealings. This resulted in what might be described as multi-layered, overlapping pyramids of people being dependent on each other… Exchange, reciprocity, and, above all, the great generosity of rulers in giving donations – these were crucial features in the ideology of domination. As a consequence, it hardly ever happened that those in power had to approach commoners; it was rather the latter who sought contact with the political authorities. Closely related to this state of affairs was the perception of the powerful as protective benefactors. This is reflected in the semantic underpinnings of bugabire, which implies both the verbs ‘to rule’ and ‘to give’. Thus, it is argued that in popular consciousness, ‘the exercise of power is virtually synonymous with gift-giving’ (Lemarchand, 1994:13).

There is thus a paradox underpinning precolonial social relations. On the one hand, inequality was ‘the precondition of social exchange and the prime motive for seeking the protection of a superior’. On the other, the basis of this relationship between patron and client

‘was utilitarian and sentimental, juridical and moral’ (Lemarchand, 1994:13). As Lemarchand

(1994:13) demonstrates, a number of proverbs reveal the positives resultant from this social ontology: ‘Amasabo arakize (‘Dependence makes one wealthy’); Udsavye ntakira (‘He who does not have a protector will never get rich’); Amasaka ada ku masabo (‘Sorghum grows in the shade of subjection’)’. Force, it should be noted, was also involved in the patron-client relationship at the level of political authorities at the top of the pyramid. Nevertheless, the terms 129

of exchange between patron and client were more flexible and negotiable than the ubuhake system in Rwanda, particularly its extension under colonial rule (Lemarchand, 1994:13).

The presence of several mediating social institutions that checked the royal power of the ganwa and the mwami also distinguished Burundi from Rwanda, including the abishagantahe and the banyamugamba. The abishangantahe were elders who served a judiciary role as arbiters of conflicts and as intermediaries to central power:

Part judge, part ombudsman, part moral interpreter, abashingantahe held their status by their respect in the local community and they often served to articulate local concerns. Although commonly (though not exclusively) Hutu, they were fully recognized within the Burundi political system in a way unknown – even adamantly opposed – in Rwanda under the Nyiginya dynasty (D. Newbury, 2001:274; see also Rutake and Gahama, 1998:83). Such is the strength of the memory of these figures in precolonial social cohesion that there have been many recent discussions about their reintroduction and ‘modernisation’ in situations of transitional justice (see Nicayenzi, 2002; Nindorera, 2003; Ingelaere and Kohlhagen, 2012).

Hutu people could also occupy another major symbolic and administrative role – that of banyamabanga (keeper of the state secrets). The banyamabanga were custodians of rituals at the royal palace whose power tended to extend beyond that of the mwami and the ganwa. Their influence, as well as the abashingantahe, was significantly curtailed under the colonial reorganisation of society, especially due to the Belgian aversion to certain traditional institutions that they conflated with witchcraft and by which they felt threatened (Gahama,

2001:28-29).

INTERMEDIATE REFLECTIONS

In sum, the precolonial social and political situation in Burundi was distinct from that of Rwanda. In the former, the mwamis’ political influence was subordinate to their religious and social roles; the political sphere belonged instead to the aristocratic class of ganwa. This class presided over both Tutsi and Hutu subjects, which were shifting terms used to denote 130

relationships between different social actors. As such, at the royal court, both Hutu and Tutsi were influential. Furthermore, through the abashingantahe system, the whole population was drawn into a common judicial practice; though imperfect, justice was not arbitrarily imposed from the court or by the ganwa but, through the institutions of abashingantahe, was based upon reference to a transcendent social order represented by the monarchy (Prunier, 1995a:129;

Gahama, 2001:302).

Although the relational quality of the Hutu-Tutsi distinction was constituted by inequality, and the terms were loaded with status and class differentiation, the capacity for mobility and social distance between the two groups in Burundi was greater than in Rwanda; identity was more ‘fluid’ prior to the colonial encounter (Uvin, 1999:255). The entanglement with Rwanda in the administrative union of Ruanda-Urundi, as we shall see, thus had greater

‘solidifying’ effects in Burundi than in Rwanda. However, this solidification was not total and

‘ethnicity’ did not have the same domineering effect in Burundian society as Rwanda until after independence.

Paradoxically, the cyclical fragmentation of precolonial kingship in Burundi, which rendered it particularly weak in the colonial encounter, led it to become a symbol around which national unity could be sustained (Lemarchand, 1977a:94). This can be traced to its largely symbolic role in precolonial social development, sharply distinct from the imperial conquest of the Nyiginya kingdom in Rwanda. As Gerard Prunier puts it, ‘although a distinct social domination of Hutu by Tutsi was evident, the cohesive nature of what could without anachronism already be called a nation-state, was stronger than the divisive potential of its social structure’ (Prunier, 1995a:129).

In short, it suffices to say that the ethnic categories of Hutu and Tutsi were not created ex nihilo by the German and Belgian colonisers, but neither were they fixed hereditary realities. 131

Ethnogenesis, and differential trajectories of identity formation in Burundi and Rwanda, were linked to processes of centralisation and fragmentation. In the more centralised Rwanda, they were institutionalised in various ways as the state centralised and expanded its influence into outlying regions. The monarchy was a symbol of sharp social contestation. In Burundi, Hutu and Tutsi were more ambivalent terms related to more informal social and cultural practices.

The crown played a more symbolic role, transcending the field of cyclical political struggles between ganwa factions.

We have seen, then, that the precolonial development of Burundi and Rwanda in comparative perspective demonstrates a number of important divergences. These were trampled over in the formation of an administrative union of Ruanda-Urundi. If it was seen to have developed in any sense differently from Rwanda, Burundi was perceived to be a sort of degenerate version. I now turn to the experience of colonial modernity in both countries. 132

CHAPTER SIX – THE COLONIAL ENTANGLEMENT, 1905-1945: THE RACIALISATION OF TRADITION.

In this chapter, I develop the claim that the experience of colonialism represents a specific ‘route to and through modernity’ (Therborn, 1995:5-7, 2003, 2011:54-76). In the cases of Burundi and Rwanda, colonialism was of a specific type, namely indirect rule, which can be broadly defined as a policy of collaborating with, bolstering, and ruling through traditional or customary authorities (e.g. Mamdani, 1996, 2012). In this sense, this chapter is an analysis of an experience of modernity that was imposed by indirect colonial rule onto the historically and geographically specific social and political circumstances that I elucidated in the previous chapter. This experience of colonial modernity was inaugurated towards the end of the nineteenth century and was thus heavily inflected with racial theories of the age. Subsequently,

I argue that the traditional bases of precolonial social organisation did not all ‘melt into air’ during the colonial encounter, to borrow the familiar term of Marx and Engels (2004 [1848]:7).

Rather, the reconstruction of society, especially in Belgian-led reforms during the 1920s and

1930s, entailed the solidification of selective aspects of tradition along lines of race.

Thus, the relationship between modernity and tradition in the experience of colonial modernity problematises two common conceptualisations. On the one hand, it departs from ideas of ‘detraditionalisation’ such as those presented in the works of Anthony Giddens (1990) and Ulrich Beck (with Beck-Gernsheim, 2002). Here, the movement from traditional to modern societies involves a process of individuation, a rejection of traditional sources of authority, and an increasingly personal reflexivity. A similar argument is presented, albeit in a very different idiom, in some forms of postcolonial and decolonial theory whereby colonial states swept away 133

the traditional bases of precolonial social relations tout court in the unfolding of ‘processes of colonialism, enslavement, dispossession and appropriation59’ (Bhambra, 2014:143).

My argument in this chapter also challenges the implicit notion in the multiple modernities paradigm (Eisenstadt, 2000) whereby tradition forms a relatively stable foundation or ‘cultural programme’ upon which alternative institutional arrangements of the cultural programme of modernity are erected (see critique by Wagner, 2014:295-296, 2015). What this does not capture is how we might conceive of the ‘invented’ (Hobsbawm and Ranger, 1983) and, crucially, racialised aspects of tradition and their relation to colonial powers in the practice of indirect rule as well as to precolonial elites. To varying degrees in postcolonial Burundi and

Rwanda, tradition did not constitute some value-neutral repository that could be borrowed from for the benefit at all. As I shall demonstrate in chapters seven and eight, the racialised forms of

Burundian and Rwandan traditions have been operative through their ‘effective history’

(Gadamer, 2004 [1975]:300-301) throughout the postcolonial period, and have been the site of intense and traumatic social conflict.

To begin with, I present a theoretical elaboration of the modern colonial state. In particular, I refer to Bauman’s (1987, 1989, 1991) notion of the ambivalence-eliminating and order-building ‘gardening’ state and its tension with indirect colonial forms of preserving difference. In the cases of Burundi and Rwanda, these forms of difference were informed and framed by a racial framework elaborated by the early European explorers of the region, known as the Hamitic Hypothesis. I then situate colonial rule in the country, first considering the establishment of the German colonial state before discussing the Belgian takeover and the

59 This notion is especially prominent in Latin American decolonial theory. As José Maurício Domingues (2009:114) says of the work of its most representiative theorist, Walter Mignolo, much decolonial theorising presents: ‘(1) a reductive view of modernity, in which only domination appears as relevant and no ambivalence is allowed to seep through; (2) the idea that only what is not modern – or is at least in an ambiguous relationship with modernity – is valuable in Latin America. This view is a sort of inverted mirror of modernisation theory, with serious cultural and political consequences’. 134

League of Nations mandate received in 1925. From this point onwards, Burundi and Rwanda were entangled together in an administrative union, Ruanda-Urundi, and the colonial administration embarked on a number of reforms. I discuss the effect of the shaping of colonial policies on predominantly racial grounds on the identities embedded within dynamic histories of precolonial state formation outlined in the previous chapter. In a concluding section, I claim that this formative experience of modernity would come to possess a great significance both in movements towards autonomous interpretations of modernity and in postcolonial mass violence, which are the themes of the following chapters.

THEORISING THE COLONIAL STATE

Before exploring the specificity of Ruanda-Urundi, I want to make some more general thematic claims about the modernity of the colonial state in its indirect form. I have suggested that modernity is best conceptualised as a time-orientation open towards the future. The colonial state embodies this future-orientation. As the historian of Africa, Bruce Berman

(1984:165), stated:

It is important to recognize that it represents one of the most consequential modern efforts to modify or create entire social structures. For a large portion of humanity it continues to be of enduring importance in defining their socio-economic and political circumstances. It should not surprise us, however, that colonial projects of social engineering did not succeed as planned. While the actual practice in most colonies was far less coherent and intellectually elaborated than the schemes of colonial theorists, colonialism was nevertheless the most conscious and deliberate aspect of capitalist imperialism. I suggest that an effective way of theorising the colonial state is through the modification and extension of Bauman’s metaphorical notion of the ‘gardening’ culture of modernity.

For Bauman, a core principle of ‘solid’ modernity is that it implied compulsive order building and was concerned with eradicating ambivalence and indeterminacy. This order would not occur naturally; it had to be created (Bauman, 1997ː115). Bauman thus conceives of modernity as ‘a time when order – of the world, of human habitat, of human self, and of the 135

connection between all three – is a matter of thought, of concern, of a practice aware of itself’

(Bauman, 1990ː163; see also Tester, 2004:115).

The practice of imposing order was likened to the job of ‘gardening’. The modern nation-state thus operated with a gardening stance, ‘cultivating’ or ‘growing’ its members into optimum producer-citizens. The gardening state ‘split the population into useful plants to be encouraged and tenderly propagated, and weeds – to be removed or rooted out. They put a premium on the needs of the useful plants (as determined by the gardener’s design) and disendowed the needs of those declared to be weeds’ (Bauman, 1991:20). Those who defied classification, now cast as ‘weeds’, required special treatment in order for the garden to flourish.

Assimilation, containment and (most extreme) extermination were modern order-building strategies (Bauman, 1989). All this was directed towards realising a blueprint for a perfectly designed society. The design, Bauman states, was ‘presumed to be dictated by the supreme and unquestionable authority of Reason’ and ‘supplied the criteria to evaluate present-day reality’

(1991:20). It joined ‘legislating’ intellectuals and political rulers in an uneasy alliance (Bauman,

1987; see also Smith, 1999:141-142).

There is an elective affinity between the metaphor of gardening and the establishment of colonies. In his Invention of Africa, Valentin Mudimbe (1988:1) reminds us that ‘colonialism and colonisation basically mean organization, arrangement. The two words derive from the

Latin word colere, meaning to cultivate or to design’. There is a double meaning to the gardening stance concerning the colonies in this sense. Firstly, they were deemed unkempt wildernesses in need of taming and transformation into an ‘orderly society, akin to the precepts of reason’ (Bauman, 1991:20). Secondly, the term can refer to the literal cultivation of these 136

gardens, and the transformation of their inhabitants into labourers to this end60, for the purposes of enriching the metropole. As such, modern colonial-imperialism not only entailed a conquest of societies and polities, but also formed part of the modern drive to conquer and overcome nature (Bauman, 1995:200; Osterhammel, 2014:828).

As Tilman Schiel states, the colonial state was ‘a pure, unmitigated type of gardening state’ which rejected ‘any ideas of citizenship and civil rights for the ‘natives’ as obstacles to the aim of transforming these territories into well-ordered, rationally operated plantations’

(Schiel, 2005:82). Colonial states, unlike nation-states, were not concerned with fashioning ethnically homogenous populations. Indeed, indirect colonial rule tended to entail the preservation or even invention of forms of difference (Comaroff, 1998:329-330, 2002;

Osterhammel, 2005:51; Randeria, 2006:60 Mamdani, 2012:1-2). Where direct rule tended to posit a racial bifurcation between the political identities of settler and native, indirect rule worked instead to fragment ‘natives’ into different ‘ethnicities’ (Mamdani, 1996, 2001:23). It involved collaboration with and strengthening of indigenous authorities, and utilised existing customs and practices provided they were compatible with the economic rationalisations of colonial rule. To paraphrase John Cormoroff (1998:329), the solicitation of the labour of the natives entailed transforming ‘savages’ into proletarians, whilst the political practice of indirect rule operated according to the grammar of cultural diversity and racial inequality.

60 Both orientations were expressed by Henry Morton Stanley, spoken truly in the language of the gardener, on his expedition to central Africa prior to the establishment of King Leopold II’s Congo Free State: ‘In every cordial-faced aborigine whom I meet I see a promise of assistance to me in the redemption of himself from the state of unproductiveness in which he at present lives. I look upon him with much of the same regard that an agriculturalist views his strong-limbed child; he is a future recruit to the ranks of soldier-labourers. The Congo basin, could I have but enough of his class, would become a vast productive garden’ (in Renton et al, 2007:20). These orientations often worked in tandem – the ‘civilising mission’ as ideological legitimation for imperialist economic expansion and enrichment – but they could also exist in tension with one other. As Comaroff (2002:122) has argued, colonial- imperialism is intertwined with, but not reducible to, the historical development of capitalism. 137

The reasons for this were quite practical. Colonial explorers and early administrators identified what seemed to resemble states whose removal and replacement would have required considerable work, resources and likely violence. Coupled with these states’ level of precolonial development, this meant that indirect rule was particularly suitable in the interlacustrine area of Africa (Fieldhouse, 1991:117-118; Gahama, 2001:42). As the development of the precolonial states of Burundi and Rwanda had before produced political and social identities, so the form and institutions of indirect colonial rule now shaped the identities of its inhabitants. Under Bauman’s influence, I suggest that it solidified them, resulting from the central tension between the instrumental, biopolitical, ‘gardening’ orientation of the colonial state, and its romantic valorisation and preservation of difference in the form of ethnic and racial identity. The principal vector of solidification in Burundi and

Rwanda, as I shall go on to discuss, was racial identity and its institutional expression.

Colonial social engineering by no means always lived up to the ‘high-modernist’ (Scott,

1998:6) blueprints. A defining feature of colonialism is that the attempts to impose order, in the mode of the ‘gardener’, often had disorderly and destructive consequences that called for improvisatory action61. Furthermore, there were certainly liquefying (Bauman, 2000) elements of colonial rule. These include the globalisation of the capitalist mode of production, the wage- labour relationship, the introduction of new ‘Western’ needs and desires, and the propensity of capital in general towards the creative destruction of ‘fixed, fast-frozen relations, with their train of ancient and venerable prejudices and opinions’ so vividly depicted by Marx and Engels

(2004 [1848]:7). ‘Barbaric’ social and political traditions deemed to be incompatible with

Western standards of civilisation were also swept away.

61 The violent suppression of anti-colonial rebellion has to be situated here, as an order-maintaining attempt to attain mastery over a situation resultant from the disorderly consequences of earlier order- building implementation. This ties in to my previous discussion of the improvisatory intentionality of genocidal violence (see pages 27-34, 42). 138

Nevertheless, these liquefying elements were not irresistible and total in their effects; liquefaction and solidification operated in a relationship of mutuality and entanglement with one another. Indeed, Bauman (1991:10-17) recognised how efforts to impose order, to make solid, necessarily created more ambivalence and disorder, which in turn had to be subject to further solidifying, order-building actions. This process ultimately gave modern political formations, including the colonial state, their dynamism (Bauman, 1991: 14).

THE HAMITIC HYPOTHESIS AND EUROPEAN EXPLORATION

Owing to the natural bulwark of wetlands and mountains as well as their bellicose reputation62, Burundi and Rwanda were relatively isolated from outside incursions such as the

Arab slave trade. As a result, the first sustained European contacts with them began later than other confrontations, in the last decades of the nineteenth century. As sites of physical

European exploration, they actually followed the ‘scramble’ for African territory among the

European states, even though they were situated in a region that had long been a subject of

Occidental fascination and fantasy (see Moore, 1901:355; Shantz, 1922; Louis, 1963:103). As put in an overtly chauvinistic text from the Belgian office of publicity a few years before independence:

Up to the start of the 19th Century, Central Africa was ‘terra incognita’, an unknown, mysterious land. The old maps showed this part of the Dark Continent as a large white patch with, here and there, a few vague marks denoting regions, forests, rivers, mountain ranges. The ‘Mountains of the Moon’ already mentioned by Ptolemy stand serried in the very heart of Africa, and haunt the dreams of explorers (de Meyer, 1958:13; see Chrétien, 2003:202).

62 This reputation particularly characterised early European encounters with Burundi (Louis, 1963:101). As a botanist called George-Francis Scott Elliot (1896:264-266) wrote, ‘my journey through this country was the most dangerous and difficult attempt that I have ever undertaken … The character of this country has had a curious effect on the people. It is so fertile that the population is extraordinarily dense. This is, perhaps, also due to the fact that no Arabs have ever obtained entrance to Mwesi’s and hence no slave raiding has occurred. Moreover, the manner in which it is cut up by valleys and swamps has very much isolated the different villages. Each is perched on its own ridge and, through the struggle for existence, is perpetually fighting with all its neighbours. The boys and young men of each village form a standing army perpetually on the watch and ready for anything’. Furthermore, the first station of the White Fathers missionaries in 1879 in Rumonge was destroyed after its inhabitants were massacred in 1881 (Shantz, 1922:330). 139

In this broad, regional sense, John Hanning Speke and Richard Burton first entered the region in the late 1850s, in their search for the sources of the Nile (see Speke and Burton, 1858;

Baker, 1962). Henry Morton Stanley’s infamous encounter with David Livingstone at Ujiji on

Lake Tanganyika occurred in 1871 (Driver, 1991). Carl Peters, founder of German East Africa, began his mission in 1884 (Iliffe, 1969:11). Oscar Baumann became the first European to enter

Burundi and Rwanda in 1892 (Chrétien, 1968). He was followed by other German explorers like Count Götzen and Wilhelm Langheld (1894), General Lothar von Trotha (1896), who would later instigate genocide in German southwest Africa, Hans Ramsay and Richardt Kandt

(1897), Heinrich Bethe (1898), and the Duke of Mecklenburg (1907) (Sholz, 2015:32).

These meetings, and others in between, roughly correspond with the beginnings and middle stages of what Moses (2002) has called the ‘racial century’. This period stretches from

1850 to 1950 and refers to the ‘competition between rival projects of nation-building and

‘people-making’ (that is, the fashioning of ethnically homogenous populations domestically) that culminated in the Holocaust of European Jewry and other racial minorities in the 1940s’

(Moses, 2002:33-34). It is significant that the colonial entanglement in the Great Lakes, and the establishment of colonial states in the region, begins during this ‘Golden Age of Racism’, the culmination of a concept of race which constitutes ‘a fully evolved system for the hierarchical ranking of humanity, from superior white to inferior black’ (Lentin, 2008:491; see also Delanty, 2013:191-192).

The racial theory employed in the Great Lakes region, the Hamitic Hypothesis, was emblematic of this. A mixture of theological, biological and anthropological thought

(Campbell, 1999:58), it provided a pseudo-scientific justification for domination and was thus incorporated into the institutional structures of the colonial state. This was the idea that all signs of civilisation in Africa can be traced to semi-Caucasian descendants of Ham, son of Noah, who originated in the Middle East (Chrétien, 2003:50). This notion originates in the writings 140

of the British explorer John Hanning Speke. Speke was struck at the Karegwe royal court in

Northwest Tanzania by the populace of the court, ‘their fine oval faces, large eyes, and high noses, denoting the best blood of Abyssinia’ (Speke, 1864:203).

In 1894, in Rwanda, the German explorer Count Adolf von Götzen clearly demonstrated his allegiance to Speke’s ideas, speaking of ‘great migrations of Hamitic peoples who came from Abyssinia and Galla countries’ who rule over a ‘great mass of the local population, the agriculturalist Wahutu, a tribe of Bantu Negroes who have been living here since time immemorial’ (in Chrétien, 2003:71). The supposed origin of the ‘civilising’ Hamitic or Galla countries in Abyssinia or Ethiopia is of particular theological significance. Christianity was established in Ethiopia long before the penetration of the interior of the continent

(Therborn, 2011:42). Hamitic or Galla peoples of Ethiopian descent – sons of Ham – were deemed a civilising-invading force in nineteenth century racial historiography precisely on account of their presumed status as an ‘outpost of Semitic civilisation’ (Levine, 2000

[1974]:18).

In this schema, the Hamitic Tutsi were a ‘great immigrant race that appears to have come down in distant times from the north’ (Jack, 1914:245) to rule over the Bantu ‘Bahutu and dwarf Batwa, who are the slave tribes or working classes’ (Barns, 1923:40-41). These identities fused biological and cultural features and developed over the duration of the colonial encounter, incorporating developments in anthropometrics and pseudo-scientific disciplines like phrenology (Basaninyenzi, 2006). They were also suffused with moral judgements

(Gahama, 2001:279). A Belgian doctor discussed in 1948 how the Hamitic Tutsi ‘are 1.9 metres high’ , are ‘slender’ and ‘possess straight noses, high foreheads, thin lips’ and discerned in them a ‘certain refinement’ that – in a way that anticipates some of the genocidal discourse of the postcolonial period – masked ‘a sense of treachery’ (Dr. Jules Sasserath, in Chrétien,

2003:72). This interpretation was influential in scholarship well into the twentieth century. 141

According to Charles Seligman, ‘the Hamites were… the great civilising force of black Africa’ and were ‘quicker witted than the dark agricultural Negroes’ (Seligman, 1966: 8, 100).

Contained in this hypothesis was the assumption that separate groups of Hutu and Tutsi had over time migrated to the region to join the indigenous Twa. However, as we have seen above, such a ‘sedimentary’ view of settlement is untenable with an appreciation of the historical development of these labels, their changeability over time, and their links to political centralisation and social hierarchy, (Vansina, 2004:198). The early explorers also assumed a

‘denial of coevalness’ (Fabian, 1983), particularly in the idea that the Burundian and Rwandan societies that they encountered (which we have established were organised according to different logics) were representative of static, feudal societies, organised according to inherent characteristics of the racial groups there present (Chrétien, 2003:170-171). In this model, Tutsi fitted the bill as a carrier of certain elements of civilisation. Burundi and Rwanda were thought to have existed in a timeless, ethnographic present, only infused with developmental capacity because of outside forces that were ultimately related to Europe.

In his analysis of ambivalence in modernity, Bauman (1991:102-159) discusses figures who defy classification, pinpointing the case of the German Jews who spoiled the easy dichotomy between friend and enemy. The colonial encounter in this sense was fundamentally ambivalent (Schiel, 2005:82). In the Hamitic Hypothesis, Tutsi were gifted in leadership, intelligent and beautiful, on the one hand, but on the other they were snobbish, sly and calculating. The Hutu were timid, dishonest and lazy, but were considered jovial and friendly peasants. The Twa were capable of devotion and attachment, but this was given an animalistic connotation (Rutake and Gahama, 1998:85)

The Hamitic Hypothesis has been ‘relegated to European fantasy’ (Eltringham,

2006:426; see also Sanders, 1969; for examples of scepticism dating back as far as the colonial 142

period, see Chrétien, 2003:73). I do not intend to reiterate in any more detail, as many have already, that the Hamitic Hypothesis with its primordialist view of antagonistic races was wrong; indeed, that it is objectively incorrect is of less importance than its effective force in history as an idea (Weber, 2001 [1905]:48; see also Said, 1978:117; Hall, 1992:205). As

Terence Ranger suggested (1983:212), the ‘invented traditions of African societies – whether invented by the Europeans or by Africans themselves in response – distorted the past but became in themselves realities through which a good deal of colonial encounter was expressed’.

What I intend to do here is to see how the Hamitic Hypothesis was used to restructure institutions and solidify social relations as ‘races’ in Ruanda-Urundi. I will assess how these restructurings undermined the modern commitment to societal and individual autonomy as well as pre-empting the supposed rupture from tradition that is characteristic of modern societies.

SITUATING COLONIAL RULE

I have thematised some general features of the colonial state, and introduced the racial schema of the Hamitic Hypothesis. At this stage, I want to situate colonial rule in Burundi and

Rwanda more specifically. As I have argued, colonial empires were defined along national lines, and as such possess distinct characteristics. There are, in short, varieties of European colonialism, each with particular conceptions of the ‘civilising mission’: French, British, and so on (see Osterhammel, 2014:826-831). Burundi and Rwanda were first administered as part of Germany’s African empire, before forming part of Belgium’s following World War One.

It is well established that German self-images and historical development were defined in counterpoint to those of neighbouring European nation states. Before it came to refer to a

German regression into barbarism, the Sonderweg thesis had in the nineteenth and early twentieth century referred to the idea that German history had taken a special path which deviated from ‘normal’ Western modernisation processes (Joas, 1996:48; Varcoe, 1998). This 143

was emblematised in the expression kultur, with an emphasis on ‘inner values’ of the spiritual substance or soul common to a collectivity, and its pursuit of ultimate goods or ends. This was counterposed against the cold and ‘exterior’ utilitarianism and instrumental rationality of

British, French and US zivilization (Brecht, 1959:350; Elias, 1994:3-45; Harrington, 2016:3).

Materially, German unification had occurred at a later stage than other European powers, and industrialisation and urbanisation existed alongside a strong military-aristocratic ethos

(Dunning and Mennell, 1998). Osterhammel (2014:829) situates German colonial-imperialism outside Europe in this context:

Until 1884 the Germans had no overseas colonial empire in which they could carry out ‘cultural work’ (as it was called in those days). The German idea of education in the Classical and Romantic periods was a programme of personal self-cultivation, not without a strong dose of political utopianism. For lack of barbarians in the flesh, the civilising process turned reflexively inward at an individual level. But once Germans had the chance to take part in a grand civilizing project, they did so with particular relish. We shall see that the ramshackle rule of Burundi and Rwanda in the late nineteenth and early twentieth century failed to live up to this grandiose billing, partly due to their low prestige and profitability against the neighbouring areas of Congo and Tanganyika. The duration of the

German presence was also too short-lived to make much impact. Germany’s overseas colonies were seized at the Treaty of Versailles and redistributed, fatefully paving the way for the imperial expansion of the Reich in Europe (Outhwaite, 2008:9; Gerwarth & Malinowski,

2009:290).

Belgium also had a distinct trajectory from other European colonial-imperial powers.

Like Germany, Belgium was a fragmented society in the early twentieth century at the beginnings of its colonial period. The creation of Belgium can in part be imputed to the interaction of larger European imperial states like Holland, Germany and France. Belgium emerged following a successful revolt in the Dutch southern provinces. It created a binational country, split between the French-influenced south (Wallonia) and the Dutch-influenced, 144

Flemish-speaking north (Flanders). Upon this divided base, a unitary state was established with

French as its primary language (Gann, 1979:3-51; Young, 2015:107). Throughout the modern period, Belgium has been a theatre for intra-European conflict63 .

I suggest that three significant results for the expression of Belgian colonial- imperialism follow. First, a status as a latecomer to the plane of European national power relations; the idea of ‘catching up’ with other Western powers undoubtedly fed into the ambitions of King Leopold II in his annexation and brutal rule of the territory named the Congo

Free State (Hochschild, 1999). Belgian imperial ambitions were in the first instance an aristocratic endeavour viewed with ambivalence by the wider population (Ewans, 2002).

Following an international humanitarian campaign to end the barbarism of the Congo Free

State, the Belgian government assumed power. Enthusiasm for the Belgian overseas colonies grew incrementally. By the end of the 1950s, the personnel in the administration of Belgian

Congo, to which Ruanda-Urundi was attached, numbered ten thousand, not including the six thousand European missionaries and staff of the privately owned mines and plantations (Young,

2015:108).

Secondly, the extension of Belgian influence in Africa can be seen as an attempt to forge a stronger national societal self-understanding and assuage the ethnic schisms at home.

Thirdly, and relatedly, the secessionist heritage of Belgium and its socio-political rifts also had an impact on the sorts of alliances that emerged in the colonial entanglement. I will pick up this theme in the next chapter, particularly in reference to a generation of missionaries who entered Ruanda-Urundi following the Second World War, having been marginalised in

63 It is worth repeating the incantations of J. P. Arnason (2003:349-350) at this point. Such historical trajectories demonstrate the importance of pluralising the idea of Europe – its regional patterns, historical dynamics and civilisational constellations – alongside the critique of Eurocentrism. 145

Belgium and animated by the trauma of European barbarism, and who identified with Hutu nationalism. At this point, I turn to the establishment of colonial states.

THE FOUNDING OF THE GERMAN COLONIAL STATE In the previous chapter, we saw how both Rwanda and Burundi were encountered in distinct states of political crisis. This had a large effect on how they were governed and the sorts of alliances that were produced between local political actors and the colonial administration. Early on, in Rwanda the colonial powers more often than not sided with the monarchy and facilitated its expansionary aspirations (C. Newbury, 1988:53), while in the more fragmented Burundi they were often used by and played off against competing and conflicting princely factions (Botte, 1985:67). This disorder was exacerbated by a ‘panorama of catastrophes’ including volcanic eruptions, the spread of ailments like sleeping sickness, famines, and the general ecological disorder that was tied to European arrival, as well as the precolonial political situations (Chrétien, 2003:223; see also Botte 1985, 1985a). The ease with which colonial modernity was established was linked to this disorder as well as to the superior weaponry of the colonisers.

Despite their precolonial differences, Burundi and Rwanda were each subsumed into more encompassing geographical regions based on the already existing colonies of the

European powers (in this case, Germany and Belgium), and would ultimately come to be administrated as one unit under Belgian rule as Ruanda-Urundi. In the case of Germany, the area comprising Burundi and Rwanda was co-opted into German East Africa, an area that also included present-day Tanzania and some of Mozambique, which stopped at the boundaries of the Congo Free State to the West and the British protectorate of Uganda to the North. Following the East Africa campaign in World War I and the Treaty of Versailles, Burundi and Rwanda were appended to the Belgian Congo, whereupon the administrative territory of Ruanda-Urundi 146

was established following a League of Nations mandate in 1924. Ruanda-Urundi became a UN trust territory in 1946 (Louis, 1963; Gakuba, 1991).

The German colonial presence in Burundi and Rwanda was established in 1890 in the form of military occupation. It was later, in 1906 in Burundi and 1907 in Rwanda, that colonial policy was formed on the basis of indirect rule, owing to the centralised precolonial political systems and the fact that displacing the monarchy in each area would have led to considerable local resistance (Lemarchand, 1970:49). Captain von Grawert (in Lemarchand, 1970:40) of the

German administration in 1905 stated that:

The ideal is: unqualified recognition of the authority of the sultans from us, whether through taxes or other means, in a way that will seem to them as little a burden as possible; this will link their interests with ours. This ideal will probably be realised more easily and earlier in Ruanda, which is more tightly organised, than Urundi, where we must first re-establish the old authority of the sultan, which has generally been weakened by wars with Europeans and other circumstances64. As such, the establishment of the colonial presence in Burundi was more difficult that it was in Rwanda. As we have seen, Burundi was fragmented into semi-autonomous princely regions on the eve of the colonial encounter. Mwezi Gisabo, the ruler of the kingdom of Burundi, found his power threatened by the struggles among the ganwa, and was hostile to the Europeans on the basis that they exacerbated the precarity of his position. As such, the German colonial administration was often implicated in local political struggles, notwithstanding its early commitment to non-intervention.

64 This is significant for two reasons. Firstly, it is uttered in the same year that the German administration perpetrated the first genocide of the twentieth century against the Herero in present-day Namibia. Von Grawert’s rhetoric is a far cry from that of the acting General in German Southwest Africa, Luthar von Trothar, who proclaimed: ‘The exercise of violence with crass terrorism was and is my policy. I destroy the African tribes with streams of blood and streams of money. Only following this cleansing will can something new emerge, which will remain’ (in Mamdani, 2001:12). Secondly, this discrepancy between German colonial practices can be seen in line with the points made in the previous section of this chapter; it demonstrates the difficulties of speaking of a singular ‘national’ conception of colonial- imperialism, let alone a European one constitutive of a shared civilisational project (Steinmetz, 2008:589). 147

Because of the complexity of this situation, there were often staunch disagreements among functionaries in the colonial apparatus. This was particularly notable in the clash between the non-interventionist position advocated by General von Götzen in Tanganyika – that which was generally stated in German East African policy – and the resident Captain von

Beringe’s opinion, which stressed the necessity of ‘pacifying’ the Burundian populace before effective administration could be implemented. Such clashes over colonial policy sometimes manifested in physical violence (Louis, 1963:115; Lemarchand, 1970:52-53). Finally,

Lemarchand (1970:51) notes, German policies in Burundi manifested in a period of ‘divide and rule’, ‘characterised by an attempt to ‘freeze’ the status quo in such a way as to prevent the crown from gaining a permanent ascendancy over the chiefs, and vice versa’ (see also

Lemarchand, 1977a:102).

German colonial policies in Rwanda, under the stewardship of Richard Kandt from

1907, were less prone to violent repression, mainly due to the fact that ruling through an indigenous elite was more straightforward. We have seen that Germans arrived in Rwanda during a time of interregnum. Rwabugiri, ‘the most active and conscious embodiment of the conquest/centralisation/social standardisation process’ (Prunier, 1995:23) during the nineteenth century, had recently been killed on an expedition in Kivu. His death was followed by a period of turbulence involving coups, purges and suppression, the latter aimed at preventing revolts by groups opposed to the relatively new centralisation (Lemarchand,

1970:56-57; Prunier, 1995:23-24). From the beginning, then, the European presence in Rwanda was a determining factor in reinforcing the mwamiship and supporting the court’s increasing hold over the peripheries. According to Prunier (1995:25):

the German presence was structurally essential since it inaugurated a colonial policy of indirect rule, which left considerable leeway to the Rwandese monarchy and acted in direct continuation of the precolonial transformation towards more centralisation, annexation of the Hutu principalities and increase in Tutsi chiefly power (see also Lemarchand, 1970:57; Vansina, 2004:179). 148

As we have seen, this was supported by the Hamitic Hypothesis outlined above, and it laid the foundation for Belgian indirect rule.

Even so, the German colonial administration in both territories was weak and understaffed65 (Lemarchand, 1970:56-63). The figures at the metropolitan centre seemed more concerned with the larger and more lucrative African territory of Tanganyika, even if there had been efforts to make Burundi and Rwanda economically profitable, such as the development of coffee plantations based on African labour, and plans for a railroad, which were stalled by the onset of World War One (Chrétien, 2003:259).

Moreover, German rule had often appeared messy and incoherent, and particularly in

Burundi where it found itself amidst complex factional struggles, was prone simultaneously to standoffishness at some junctures and to despotic suppression at others, which undermined its claim to rule indirectly through the local administration. In a sense that vindicates John

Comaroff’s (2002:120) arguments about the nature of the colonial state, the German administration in Burundi and Rwanda was not a coherent political unit progressing inexorably toward the realisation of clear aims and ambitions. On the contrary, the actions of administrators often set in motion unplanned processes, which themselves provoked contestation over the appropriate response. German rule possessed an improvisatory character, and physical violence was often reverted to in response to the unintended consequences of prior social action as well as the resisting agency of the colonised.

65 The aloof nature of the German administration is reflected in the scant scholarship that considers it (Scholz, 2015:13). There is a large literature on German colonialism elsewhere in Africa, such as Southwest Africa or Tanganyika. There is also a sizeable general literature on German colonialism, in which Ruanda-Urundi is very often only mentioned in passing (Okupa, 2006; Zimmerer, Perraudin & Heady, 2011; Langbehn & Salama, 2011). There are some German language considerations of Ruanda- Urundi (Bindseil, 1992; Strizek, 2006), but these are few and far between. 149

THE FOUNDING OF BELGIAN RUANDA-URUNDI In an indictment of von Grawert’s ideal outlined above, an important governor in the ensuing Belgian administration, Pierre Ryckmans (1953:46), said that German rule in Burundi had left the country:

in a state of avowed bankruptcy because it had worked toward the disintegration of a kingdom whose traditions, mores and religion were unknown or ignored because in tolerating successful revolts, it had encouraged intrigues instead of suppressing them; because each blow against the prestige of the monarchy had rendered the white man more odious to the masses, attached above all to the traditions of their divine monarchy66. These condemnatory remarks about the legacy of the German colonial administration are curious considering that, at first, Belgian colonisation was more or less continuous with the indirect model that preceded it (Lemarchand, 1970:64-65; Prunier, 1995:26). It was only in

1925 that Belgium received a full League of Nations mandate over Burundi and Rwanda. They continued to be governed as Ruanda-Urundi, which was affixed to the Belgian Congo, partly justified by the potential for the sizeable population of the former to offer a labour force for the exploitation of the scarcely populated latter (Gahama, 2001:47; Young, 2008:700).

This administrative fusing was significant. Chrétien (2003:310) suggests that

‘Burundi’s fate [might] have been radically different if the country had not been falsely twinned to Rwanda under colonialism’ (see also Lemarchand, 1980:54-55). From the origins of the encounter with the regions, there was an assumption that Burundi was a sort of pathological version of Rwanda, which we might argue continues to this day. As such, it was subject to more or less the same indirect rule principles as Rwanda, namely that it should be ruled as a

66 The self-appraisal of Belgian colonialism towards the end of their rule echoes these assertions: ‘The Germans left a rather poor record of their sixteen years of administration: one cannot speak of colonisation, but at the very most, of a military occupation which was limited to pacification and security measures. The number of European military men and civil servants represented hardly more than ten units; no modification of the native social organisation was carried out: the arbitrary power of the native aristocracy was not in the least affected; finally, a census was never made, nor were taxes imposed nor economic expansion carried on’ (BCRUIPRO, 1960:71-72). 150

Hamitic pastoral aristocracy and that it was the role of the colonial state to support this indigenous power.

Also very striking was the attachment of Ruanda-Urundi to the administrative framework of Belgian Congo. Rule in Ruanda-Urundi was comprised of the same sort of bureaucratic paternalism that characterised practice in Belgian Congo following its inheritance from the Congo Free State (see Ryckmans, 1948; Dunn, 2008:1156-1157). For Joseph Gahama

(2001:49), although it did not affect the principle of indirect rule, this:

constituted a flagrant violation of the terms of the mandate, to the extent that the assimilation [of Ruanda-Urundi] into the Belgian Congo, from an administrative and legislative point of view, removed an autonomous future from the country... the legal framework thus defined, Belgium could henceforth politically reorganize Burundi to better exploit it economically. I have suggested elsewhere67 that one form of entanglement captures how processes of societal modernisation take place in contexts of inter-societal interdependencies. Not only are

Burundi and Rwanda entangled administratively as Ruanda-Urundi, but they are entangled with Belgian Congo. The colonial administrations of Belgian Congo and Ruanda-Urundi are also entangled together, as well as with the metropolitan government in Belgium itself. This entanglement had legal and administrative features, as aforementioned, but was principally economic. Ruanda-Urundi, Gahama (2001) argues, was effectively doubly peripheral, functioning as a peripheral sub-colony of Belgian Congo, itself peripheral to the Western capitalist core.

Culturally, the Belgian administration of Ruanda-Urundi was characterised by a staunch paternalism. Continuing the indirect rule striven for and partly implemented by the

German administration, Ryckmans placed ‘legitimacy’ at the core of the Belgian colonial policy in Ruanda-Urundi:

67 See pages 73-75. 151

The cooperation of the kings constitutes an indispensable element of progress and civilisation… Without them the problem of government would remain insoluble. There are among the chiefs some who are incapable, imbecile, who will never gain authority… There are some who are irreducibly hostile and who will never accept civilisation… These chieftaincy crises are everywhere a great stumbling block of native policies. To resolve them by dismissing a bad chief and appoint in his stead one more amenable to European influences is tantamount to substituting impotence for insubordination. Legitimacy is a moral factor of incalculable importance (Ministère des colonies, 1926:63). Legitimacy, Ryckmans said, was taken to be ‘more powerful than violence’, and as such legitimate chiefs constituted the ‘only smoothly functioning organ between us and the masses’

(in Lemarchand, 1970:66).

However, the paternal instincts of the colonisers were distinct from colonial paternalism in other contexts, such as that of the British administration. Early in the history of the Belgian administration, in 1920, Minister Louis Franck suggested that: ‘The Negroes need to be introduced to economic development alone... There should be no question of affecting the very foundation of the political institutions on the pretext of inequality’ (in Chrétien, 2003:287).

Later on, when educational policies were introduced, the elites were taught to be ‘modest and restrained’ (Chrétien, 2003:287). Here, paternalism seemed to desire the eternal childhood of its subject. Love for the colonised was frequently expressed (Duarte, 1995), but it was a possessive love involving the stagnation of the developmental capacities of the Other, and ready to resort to violence when questioned.

As I suggested above, it was subsequently the case that controlling the population instrumentally and removing the traditional bases of social life was not the sole concern nor orientation of the colonial state. Its representatives were also significantly infatuated in a

Romantic sense with the colonised, animated by colonial-anthropological descriptions that fetishised tradition, like those in the Hamitic Hypothesis. In this sense, to use the phrase of

Balandier (1970:23; see also Steinmetz, 2007:2), ‘the history of a ‘colonised’ society is worked out as a function of a foreign presence’. Gadamer (2004 [1975]:275) termed this ‘Romantic 152

refraction’, and argued that it emerged alongside the rise of the historical sciences in the nineteenth century:

It no longer measures the past by the standards of the present, as if they were an absolute, but it ascribes to past ages a value of their own and can even acknowledge their superiority in one respect or another. The great achievements of romanticism – the revival of the past, the discovery of the voices of the people in their songs, the collecting of fairy tales and legends, the cultivation of ancient customs, the discovery of the worldviews implicit in languages, the study of the ‘religion and wisdom of India’ – all contributed to the rise of historical research, which was slowly, step by step, transformed into detached historical knowledge. I suggest that a defining tension of colonial modernity in its indirect form is that between the Romantic orientations of preservation and the defence of difference on the one hand, and the Enlightenment tendency towards the expansion and rationalisation of power reified in economic and political structures on the other (the tension between Romanticism and

Enlightenment is the constitutive tension of modernity for J. P. Arnason68: see Arnason, 1991;

Friese and Wagner, 2000:34-35; Adams, 2009:257). We can see this in some other statements by Louis Franck:

We must respect the existing political and social organisation, whilst improving it by eliminating barbaric practices and abuses, and preventing the oppression of the poor and the downtrodden, and try to import, in the tropics among primitive peoples, our egalitarian European concepts, allowing black humanity, in one of its most interesting parts, to evolve according to its own nature after the traditions of its own resources by making efforts to reach its originality and its particular soul. This is a wise and just policy from which there can gradually emerge a strong and fertile African race (in Gahama, 2001:41).

We will apply, in Ruanda and Urundi, a policy of colonial protectorate. This policy is based on the respect of the indigenous institutions; it uses the European as a guide and a protector; it excludes direct administration. This policy is perfectly suited for countries enjoying an old and remarkable organization, where the ruling class shows obvious political talent. But this method does not limit itself to respecting and using the indigenous institutions; it also wants to develop them in order to adapt them gradually to the needs of the colonisation and to the economic progress of the country’ (in Reyntjens, 1987:73) In the next section I look at how specific reforms of various institutions in Ruanda-

Urundi under Belgian indirect administration are expressive of this tension. As a resident in the Rwandan capital Kigali put it in 1938, the aim of indirect rule was to act as ‘a safeguard of

68 As he puts it: ‘The point is not to oppose Romanticism to Enlightenment as an alternative project, but rather to underline a constitutive polarity of modern structures of consciousness that can express itself in highly different constellations’ (Arnason, 1991:2010). 153

traditions and a brake upon their evolution’; as ‘a melting pot in which past and present tendencies [would] coalesce’; and as ‘the means whereby a progressive and progressist, yet slow and smooth, assimilation could be achieved’ (in Lemarchand, 1970:75-76 emphasis added).

THE COLONIAL RECONSTRUCTION

The transformation and rationalisation of Ruanda-Urundi that occurred under the

Belgian administration began in earnest in 1926 with the League of Nations mandate, in policies incrementally implemented up until 1936 69 . Here, the reforms allowing for development of the indirect rule imposed by the German administration were institutionalised in accordance with the ideas associated with the Hamitic Hypothesis and its attendant associations with timelessness (Mamdani, 2012:3).

These ideas, and the ordering imperatives to which they were directed, found form in endeavours to quantify the population of Ruanda-Urundi. The scientific certitude purportedly afforded by the Hamitic Hypothesis did not make measurement easy. All of the usual indicators of ‘racial difference’ did not exist here among populations who belonged to the same clans, inhabited the same collines, were subjects of the same king, and used the same language. With the aid of the pseudo-scientific disciplines of phrenology and physiognomy, the Tutsi could be measured ‘objectively’; defined as taller, with finer noses and facial structures and different blood composition, accordant with the ethno-mythological descriptions of earlier explorers. On the basis of this counting, identity cards were issued in 1935, which were later to become infamous for their role in the 1994 Rwandan genocide as the principal means of setting Tutsi people apart at road blocks around the country (Uvin, 2004:154-155).

69 These have been labelled by some as les réformes Voisin, after Governor Charles Voisin (Prunier, 1995:26). 154

Later on, the census calculated people as Tutsi on the basis of cattle ownership

(Twagiramutara, 1998:114), reflecting the strong theme in colonial scholarship of a labour split between the pastoralist Tutsi and the agriculturalist Hutu. Relatedly, a generalisation and entrenchment of the ubuhake client-patron relationship in Rwanda was also observable. In the atmosphere of generalised insecurity that colonialism brought, more Hutu sought to enter client relationships as a means of social protection from Tutsi patrons, contributing to the further hardening of social relationships.

Ubuhake also became more coercive and extended beyond its precolonial centering in the areas around the court. Nevertheless, the colonial administration ‘had a priori seen this contract as a ‘pastoral serfdom’ that for centuries had been part of a ‘feudalism of Tutsi conquerors’’. In the name of respecting ‘customary law’ and the authority of the native courts,

‘they had been persuaded to respect its authenticity’ (Chrétien, 2003:271; see also Newbury and Newbury, 2000:861). It was posited as a traditional mechanism of functional integration in Rwandan society. Tutsi patrons leased cattle to Hutu clients, and it was assumed that all parties (defined racially) entered this arrangement voluntaristically.

The ‘premise of inequality’ was said to have reigned in a more or less static sense. On the eve of independence, Jacques Maquet and Maurice D’Hertefelt (1959:217), two Belgian sociologists with links to the administration, stated:

In the ancien régime the caste system was stabilised. This equilibrium was apparently largely accepted at least in the regions of central Rwanda and central Burundi. Should it not be feared that, as the Hutu caste become more politically aware, a system of universal suffrage whose results will not be buffered by successive ballots like today will allow them to dominate the country’s political life in such a way that the Tutsi will be deprived of all power and be sacrificed? Such a fear places too much weight on political power… It may be predicted that universal suffrage will only lead to a situation of greater equilibrium between the two castes. We will see in the next chapter that this did not come to pass. Independence was expressed in the form of a social revolution that saw a reversal of power in favour of the majority Hutu and the first systematic bout of mass ethno-republican violence (Linden (1977:240) 155

At issue here is the relationship between colonial scientific knowledge, the colonial administration and traditional authorities. Newbury and Newbury (2000) argue that, following

World War One, it became expedient for the Nyigingya kingdom, colonial administrators and the Catholic Church to present an avowedly feudalist state history of Rwanda. Regional and local struggles and differentiation among the population and peasantry were downplayed. It is worth quoting Newbury and Newbury (2000:848) at length in their discussion of these themes:

The new ‘doxic’ vision of Rwandan history consolidated the administrative, court, and missionary perspective into a single secular narrative, one later to be taught in the schools, promulgated in the press, legitimated in academic works, and ‘confirmed’ in diplomatic handbooks. The essential elements of this vision stressed the homogeneity of the society, the power of the monarchy (including ethnic stratification), and the longevity of the kingdom. Rwanda was portrayed as a unitary and enduring society, completely consolidated internally and clearly demarcated from its neighbours. To be sure, both these points contradicted much of the earlier written and oral historiography of this region. Nonetheless, this historical image of a unitary ‘Rwanda’ became as central to its claims to chronological longevity, and there was a connection between the two. In this context, ‘longevity’ was a powerful factor in legitimizing ‘the state’ – the claim to time depth implicitly projected the colonial assumptions of ‘the traditional monarchy’, unchanged, into the distant past. From this, there developed an image of a highly centralized, rigidly stratified, and ancient state. From the 1930s, political issues came to focus not on region but on royalty, social issues focused not on ecological distinctions but on ethnicity, and historical issues focused not on local initiatives but on external origins. It was in the interest of various entangled institutions of power to maintain the idea of a

‘traditional’ Rwanda.

Also in Rwanda, reforms under Voisin dramatically reduced the presence of Hutu people in ruling positions, to the extent that by 1959, 43 of 45 chiefs and 549 of 559 subchiefs were Tutsi. The complex and multi-layered chief system – comprised of networks of army chiefs, land chiefs (often Hutu in precolonial Rwanda) and pasture chiefs – were streamlined in the creation of a hierarchical system of chiefs and subchiefs, almost always Tutsi, who administered a single territory and were responsible for each of the military, pastoral and agricultural duties of the chiefs in the old system (Newbury, 1988:62-64; Prunier, 1995:27).

As Lemarchand (1970:119-120) details, this rationalised form of chiefly rule significantly hardened identities and fermented grievances. It played a particularly prominent role in the memories of colonial rule as experienced by Hutu and recounted during the transition 156

to independence and after. This bureaucratisation also led to serious abuses by single chiefs.

Often, tradition was appealed to as the basis of violence or force underpinning new obligations on the peasantry. Because of the force and violence that accompanied the colonial rationalisation, as Lemarchand puts it, 'the ‘civilising’ aspects of the Belgian presence made the rule of the chiefs a singularly uncivilised one' (Lemarchand, 1970:123). The harsher the chiefs, and the more animosity they attracted from their subjects, the greater was their reputation among the colonial administrators.

Furthermore, for the purpose of rationalising taxation, the structure of the uburetwa corvée system was extended to areas of Rwanda where it had not existed previously, and its functioning was also altered. Whereas it had previously aimed at the incorporation of lineage groups, who could delegate one member to work on their behalf, it now became individualised.

In line with the Hamitic Hypothesis, the Hutu ‘agriculturalists’ were forced to work under this system, often experiencing exceptional brutality (C. Newbury, 1980:103; Prunier, 1995:27-28;

Mamdani, 2001:97-98).

Despite Pierre Ryckman’s argument about the virtues of legitimacy over violence, Hutu men ‘were almost constantly under mobilisation to build permanent structures, to dig anti- erosion terraces, to grow compulsory crops (coffee for export, manioc and sweet potatoes for food security), to plant trees or to build and maintain roads... Those who did not comply were abused and brutally beaten’ (Prunier, 1995:35; Mamdani, 2001:95). In the words of Alexis

Kagame (Mamdani, 2001:95):

For several decades the country became a vast camp of forced work of a new type. The very notion of work came to be practically synonymous with corvée, to the point that the representatives of Authority themselves, natives as well as Europeans, understood it as such and interpreted it with this transformed nuance.

In Burundi specifically, the vast royal estates that covered the centre of the country, once run by those loyal to the mwami, were divided into chiefdoms. In Governor Voisin’s 157

words, the aim was ‘to regroup chiefdoms in such a way as to supress the dispersion of fiefs and make the administration easier and more efficient’ (in Lemarchand, 1977a:104). These were entrusted to the ganwa, notably the Bezi and Batare, and worked to undermine the monarchy further, already weakened by the actions and alliances of the German administration

(Lemarchand, 1977a:99; Chrétien, 2003:268).

Belgian colonial rule tipped the balance of power towards the Bezi. The German administration cemented the Batare in positions of power, but they were ‘eased out’ under the

Belgians, though not entirely deliberately. To quote Lemarchand (1970:312-313) at length:

For the Belgians administrative efficiency meant, essentially, the regrouping of smaller chiefdoms into larger and more viable ones. It also meant the appointment of intelligent, resourceful and, above all, authoritative chiefs. Thus, in many areas, this policy of regrouping consisted in bringing together a number of semi-independent chiefs or subchiefs under the authority of an influential ganwa, so as to consolidate several smaller holdings into a larger territorial and administrative entity. As early as 1929, the results of this policy had clearly worked to the advantage of the Bezi… This transformation, as noted earlier, was much less the result of a self-conscious, systematic pro-Bezi policy than the consequence of a de facto situation which the Belgians had to accept, whether they liked it or not. Although the official reports transmitted to the Belgian parliament by the Resident suggest a carefully planned policy, the reports from the local administrators have an entirely different ring. One gains the distinct impression that the délégués and their staff were often manipulated by the chiefs, and, while many were probably aware of what was going on in the wings, most of them unconsciously admitted that they could do little about it. It is difficult to see how it could have been otherwise, considering their lack of familiarity with the byzantine style of ganwa politics and the absence of a central ‘native authority’ worthy of the name, upon whom they could rely for assistance – the Mwami being constantly manipulated by his entourage. Whereas the Rwandan monarchy was strengthened by colonial rule, in Burundi the monarchy’s peripheral role was intensified. As we have seen, Burundi had historically been fragmented by internecine struggles among the ganwa and colonial power spurred on these factional struggles, consciously or unconsciously. These struggles defined the faultlines of competing programmes for independence, in contrast to the ethnicised independence of

Rwanda.

The members of a Tutsi elite, which included the ganwa, were nevertheless beneficiaries of colonial rule in Burundi, especially at the level of subchief. This position was bequeathed by the racial colonial reconstruction. Where Hutu people had occupied around a 158

quarter of subchief posts at the beginning of the century, none remained in these positions by

1945, whereas the Tutsi’s share rose in the same period along with ganwa (Abrams, 1995:146).

This was part of a general trend of the shaping of Burundi along particularly Rwandan lines.

As Joseph Gahama (2001:69) states:

Wrongly equating the political and social institutions with those of Rwanda as they seemed to work in the late nineteenth century, colonial rule quickly established a value scale: the ganwa, Bezi then Batare, then the Tutsi and, far behind, the Hutu. The higher categories were readily categorized under the heading of ‘noble caste’ in official reports. As in Rwanda, the function of customary authorities was rationalised, and was directed towards

‘responsible activity’, such as attending to crop extension, reforestation, road maintenance, census taking, tax collection and the administration of native courts (Ministère des colonies,

1933:75; Chrétien, 2003:268). This work mobilised the entire population and became an instrument of political subjugation and surveillance. The adult African was, in Crawford

Young’s (2008:701-702) terms, ‘a functional unit, as suggested by the customary census designation ‘HAV’ – homme adulte valide – to be harnessed to export-crop cultivation, or conscripted for mine or plantation service. Internal movement required administrative organisation’. The integration into a capitalist economy created new demands in the forms of taxation, as well as new needs. Imported products contributed to the decline of artisanal traditions (Gahama, 2001:352). Development efforts were oriented towards the development of the Belgian Congo. Major decisions were made with the productivity and profitability of the larger territory in mind (Gahama, 2001:213).

Whereas in Rwanda, because of the expediency of ruling through the Nyiginya kingdom, some customary institutions had been reinforced, in Burundi we can speak more plausibly of a melting of the solid bases of tradition. A number of customary institutions and traditional events, such as the prohibition of the sorghum festival of umuganuro, were gradually displaced after being defined as ‘barbaric’ and incompatible with European civilisation. This 159

included the diminishing of complex precolonial social institutions that had worked to mitigate social conflict, such as the abashingantahe (Chrétien, 2003:269). The pretext for the colonial reform of the colonial judicial system was an aim to abolish so-called ‘barbaric customs’. To quote Gahama (2001:302-333):

The report of the administration in 1921 insisted heavily on the necessity of assuring ‘equal justice for all, open to poor as well as rich, presenting sufficient guarantees of fairness’. The colonial administration was without doubt ignorant of the impartiality of bashingantahe… in addition to the constant confusion between [the social systems of] Burundi and Rwanda… they [the colonisers] had always believed that the chief or king had absolute power in judicial matters. So they decided to control the judgments and sentences of reform that ‘violated’ this custom… the European administration assumed the right to guarantee the authenticity of custom.

The colonial administration invented a custom, the sovereign power of the mwami, which came to replace the actual customary institution of the abashingantahe. Their disintegration led to more power for ganwa and the chiefs, who as we have mentioned became more and more able to abuse with impunity. One passive form of resistance was migration, many moving to Congo (chiefly the Kivu provinces and the Katanga provinces, for economic reasons). Some went of their own volition; others were recruited by the Belgian administration

(Gahama, 2001:371-382). More active forms of resistance came in the form of revolts (Chrétien,

1970; Gahama, 2001:383-397).

Key in the colonial reconstruction of Ruanda-Urundi, in addition to the Belgian colonial administration, were the representatives of the Catholic Church. The missionaries’ introduction of Christianity contributed to a secularisation of the institutions of kingship in both

Burundi and Rwanda (Lemarchand, 1977:75; Prunier, 1995:35). In Rwanda, there developed a hostility between Mwami Yuhi Musinga and the missionaries. However, as Lemarchand

(1977:75) points out, the Tutsi as a group identified with Christianity because it was itself identified with ‘Western technology and educational achievements, and because they fully sensed the significance of the new social forces that lay behind it they naturally felt that the 160

preservation of their traditional privileges was intimately linked to their endorsement of the new creed’. Out of this, then, developed an antagonism between the Church and the aristocracy.

Chrétien (2003:273) argues that colonial policies in Ruanda-Urundi, particularly involving the Catholic Church, effectively amounted to a kind of feudalisation; Leon Classe, head of the catholic mission in Rwanda, thought that ‘a medieval-style Rwanda should be constructed, with its Tutsi aristocracy made to rule, its Hutu peasantry made to work, and its

Church made to shed light over the lot, all which working hand in hand with civilian authorities’. His attitude on arriving in 1916 was characterised by a conviction that the country was best characterised as a Middle Ages feudal regime (Classe, 1916). The notion of traditional, feudal Rwanda, split between Tutsi lords and Hutu serfs, was a colonial invention. It was the assumption of a denial of coevalness, mixed with the racial schema that Europeans used to understand Burundi and Rwanda, that had a profound influence on the way in which material relations were shaped in the territories.

The Church stood behind the introduction of educational institutions that became the principle mediating institution in the racial recasting of the Hutu-Tutsi cleavage. ‘The first

Western-style school in Rwanda’, Mamdani (2001:89) notes, ‘was opened by the White Fathers in 1905 in Nyanza… The objective was to turn the Tutsi, the ‘born rulers’ of Rwanda, into an elite ‘capable of understanding and implementing progress’, and thus functioning as auxiliaries to both the missionaries and the colonial administration’.

By 1932, the Catholic Church had become Ruanda-Urundi’s main social institution

(Prunier, 1995:32; Gahama, 2001:241). Until the 1960s, education was contained at the primary level. Secondary establishments included the Nyakibanda seminary in Rwanda,

Mugera seminary in Burundi, and the Groupe Scolaire, founded in Astrida in 1932 to train the agents of indirect rule. Recruitment statistics show that Rwandan Tutsi trainees were almost 161

four times as numerous as Hutu until 1958 (Chrétien, 2003:286). Burundian chiefs from the

Bezi princely faction were also trained at Astrida (Lemarchand, 1977a:106).

INTERMEDIATE REFLECTIONS

On the eve of independence, in a self-appraisal of the Belgian colonial administration under a heading tellingly titled the ‘evolution of customs’, it was suggested that ‘European civilisation has brought into the country an atmosphere of general security, the use of money, rapid means of communication and generalised instruction, the principal elements of revolution of the subsistence economy’ (BCRUIPRO, 1960:31). These precepts of modernity were accompanied by the pressure of Christian religion, whose effect was the disintegrating of the

‘ancestor cult’, evoking some Durkheimian themes about the anomic effects of modernisation

(Durkheim, 2006 [1897]:ch.5). ‘Correlatively’, it said, ‘family cohesion and authority are regressing’, and ‘individual selfishness tends to minimise the customary obligations toward dependent persons’ (BCRUIPRO, 1960:31-32).

The influence of modernity here is conceptualised as a melting of the solids. But from the above analysis, we can see that, with the aid of pseudo-scientific knowledge, the meetings of the colonial powers and precolonial societies of Burundi and Rwanda also resulted in the solidification of what had previously been historically dynamic and complex societies and the hardening of social identities therein.

Under the aegis of modernisation and development, the colonial state presided over structural changes in Burundi and Rwanda that served to solidify social relations into institutionalised races, which were constructed as the inhabitants of a neo-feudalistic political order (Chrétien, 2003:285). This undermined the commitment to societal and individual autonomy characteristic of the project of modernity, as well as the idea that modernity involves a break with tradition. To do this, colonial administrators made recourse to a racial theory that 162

fused theological, biological and cultural categories. They also assumed that the societies that they met had existed in traditional forms whose historical dynamic and development could only be imputed to outside civilising forces carried by Hamitic peoples from North Africa. A fundamental tension lay at the root of the order-building imperative of the colonial state, deriving from the instrumental and biopolitical management of the population and their productivity on the one hand, and the romanticising and reifying understanding of the population and its history on the other.

In the vocabulary of my theoretical framework, it is possible to observe significant societal entanglements. Rwanda, as we have seen, was taken as the model for reforms, and

Burundi was subsequently shaped to resemble its northern neighbour; this much is clear in the administrative entangling of Ruanda-Urundi. The distinct paternalism and institutional features of Belgian rule also affected the specific articulation of colonial modernity here, as did precolonial developmental dynamics in each of the territories before the Belgians’ arrival. The colonial administration of Ruanda-Urundi was also entangled with that of Belgian Congo, to which the former supplied economic and human capital, and whose bureaucratic-legal instruments were implanted onto Ruanda-Urundi.

There are numerous points of geohistorical entanglement with global institutions. The most obvious of these concerns the integration of Ruanda-Urundi into the capitalist world- system, which created new needs among and new demands on the population in the form of corvée labour extensions and taxation. There are also the political entanglements entailed in the organisation of colonial-imperialism. In 1924, for instance, Belgium was granted a League of Nations mandate over Ruanda-Urundi. Prior to this, it had been a German colony before they were divested of it following World War One. Finally, a highly significant cultural entanglement was constituted by the influence of the Catholic Church. As against theories which equate modernity simply with secularisation, we see that Catholic missionaries were the 163

self-proclaimed carriers of modern progress and civilisation, a point observed by Goran

Therborn (2011:48). It was the missionaries who saw Tutsi people as amenable to the lessons of Western civilisation through education and, as we shall see in the next chapter, the later generation of missionaries who saw Hutu people as the downtrodden majority in need of self- emancipation. Overall:

The colonial administrators introduced new forms of production destined for export (notably coffee and cotton); they diffused money, they privileged means of taxation; and they opened schools, mainly through Protestant or Catholic [the latter in Burundi and Rwanda’s case] societies. Thus the Europeans introduced new logics of profit, money, instruction, technical expertise, and mobility – in short, new logics of individual advancement, which in principle was a break with the ancient lineage networks (Chrétien, 2003:351). These new logics and dynamics, in plenty of contexts, might have ‘melted all that was solid into air’ and constituted the turn away from tradition often deemed constitutive of modernity.

That said, this melting aspect existed in tension with the solidifying effect of colonial- imperialism in Ruanda-Urundi:

Its own past was picked up and intensively invested in clichés forged by the anthropology of the late nineteenth century, even before these societies were truly understood… A veritable scientific ethnicism – or, to put it more crudely, racism – took hold in these societies. The first generations of Western-educated elites were trained in this perspective, and they were motivated by it… Mental structures can be as concrete as economic ones. In the same way, a fundamental contradiction shaped ‘modernity’: new values opened up space for personal aspirations, but these hopes were dashed because of an official policy that locked people in a racial and supposedly traditional framework (Chrétien, 2003:352-353; see also Reyntjens, 1987). A key point here is that colonial modernity in its specific, indirect variant did not trample over and dissolve traditional society. The distinct coloniality of indirect rule lies precisely in the racialisation, and thus solidification, of tradition. This argument brings sociological questions about modernity into dialogue with analyses of colonial rule influenced by postcolonial theory (i.e. Mamdani, 1996; Taíwò, 2010). My claims in this chapter are similar to Olúfémi Taíwò’s (2010:25) notion of sociocryonics, whereby:

instead of allowing social forms to evolve as they will in the ordinary course of life … under the direction sometimes consciously and most times unconsciously of human agency, humans 164

freeze those forms and pre-empt their transformation by subordinating others whose social forms they are. Whereas Taíwò suggests that the freezing of social forms pre-empted modernity in the

West African societies he analyses, I argue that this is too totalising a summation in application to Ruanda-Urundi. Rather, the solidifying or ‘freezing’ aspects of indirect colonial rule existed in tension with ‘liquefying’ aspects.

This has implications for the notion of multiple modernities. Under conditions of indirect rule, the governance of the native was entrusted to the native authority. This was based on an acknowledgement of tradition and custom. In a sense that implies a concordance with

Enlightenment ideals, the Belgian administration framed this kind of rule in terms of emancipation and autonomy. As put by the Minister of colonies (1938-1945), Albert De

Vleeschauwer (1943:27):

Conforming to the conception that colonization must be made in the interest of the natives, the Belgians intend to pursue progressively the emancipation of their native people. By emancipation is meant the chance of conducting themselves independently of the mother country. But the Belgians believe that a durable autonomy must begin from the bottom. This is a curious statement, not least because respect for ‘cultural difference’ is patently organised around an unchanging version of culture that was undoubtedly shaped by the perceptions of Europeans. Furthermore, an autonomous modernity involving a genuine interpretive fusion with socio-cultural traditions was denied in the very principle of colonial rule. Tradition was much more a tool wielded by the colonisers for the purpose of more efficiently governing colonial populations. It also tended to serve the interests of pre-existing power elites, the ganwa in Burundi and the Nyiginya court in Rwanda.

Thus, the original contribution of this chapter to contemporary debates in social theory is in its elucidation of a problem hitherto unacknowledged in notions of multiple modernities: the experience of colonial modernity, in its indirect form, entailed a colonisation of the ‘cultural programme’ (Eisenstadt, 2000) of the colonised societies that had significant consequences for 165

those autonomous interpretations which would develop in the transitions to independence. The importance of the present chapter for the thesis as a whole is that this experience of modernity constitutes the background of its interpretation in both the decolonial and post-colonial periods

(Wagner, 2008). Autonomous conceptions of modernity detectable in the discourses of independence movements in the 1950s and 1960s, based upon the principles of societal autonomy, development and justice, became entangled with customary tradition and racial identities. These were not so much invented ex nihilo by European colonial powers, but solidified and frozen by successive colonial powers in institutional reforms modelled after colonial knowledge.

166

CHAPTER SEVEN – TRAJECTORIES TOWARDS INDEPENDENCE, 1945- 1965: MULTIPLE ‘SOCIETAL SELF-UNDERSTANDINGS’

‘Reality’ always stands in a horizon of desired or feared or, at any rate, still undecided future possibilities. Hence it is always the case that mutually exclusive expectations are aroused, not all of which can be fulfilled (Gadamer, 2004 [1975]:112).

How long shall we have to wait until our injustices are redressed?… Until the Hutu no longer has the soul of a serf. For that he must be reborn (From Saverio Naigiziki’s L’Optimiste, in Lemarchand, 1970:133).

I see development, I see concerned brothers, I see men who are questioning themselves, I see men arguing about progress (From Michael Kayoya’s Entre Deux Mondes, 1971). --

It is clear from the conclusion of the previous chapter that the colonial state in Ruanda-

Urundi, in its successive German and Belgian articulations, was not simply a steam-roller which completely destroyed precolonial society to create a tabula rasa, whether through an inherent propensity to exterminatory violence, or through the dissolution of custom and tradition through the introduction of the dynamics of modernisation. I argued that this version of events fails to grasp the specificity of the relationship between modernity and tradition in the practice of indirect rule. In particular, the Belgian administration, influenced by a schema that assumed the existence of static, racially-stratified ‘traditional’ societies, stalled and solidified precolonial dynamics by ruling indirectly through this racialised tradition.

At the same time, the notion of the colonial state as steam-roller also underplays the creative agency of those subject to colonisation, both in the sense of their resilience and resistance to it and in the sense of how colonial administrations became entangled in local social-historical dynamics. From the beginning, Burundians and Rwandans resisted the intrusion of colonial powers in various ways. As acknowledged in the previous chapter,

Burundi had a fierce and belligerent reputation among early explorers and Swahili Arab traders, and there were localised revolts against colonial rule throughout the first half of the twentieth century (Chrétien, 1970; De Juan, 2015). Rwandan resistance included the Nyibingi rebellion, 167

led by the infamous Muhumusa, a liberation movement that emerged in the north-western Kiga region of Rwanda, directed against the Nyiginya kingdom and their European enforcers that

Jan Vansina revealed to be coloured by pre-racialised anti-Tutsi animosities (Chrétien, 1972;

Freeman, 1974; Rutanga, 1991; Vansina, 2004:137).

We also saw in the previous chapter how the colonial powers, often unwittingly, became embroiled in political struggles at the centres of the courts in both countries In Burundi, the colonial administration became entangled in the complex and shifting landscapes of power struggles between ganwa, ultimately weakening the political power of the Burundian monarchy

(D. Newbury, 2001).. In Rwanda, they became accomplices in the expansion of the Nyiginya court and helped to suppress resistance from less-integrated territories such as those in the

Northwest of the country who perceived the expansion of the court as a kind of imperialism

(Lemarchand, 1970; Chrétien, 1997:157; C. Newbury, 1998). As I will go onto discuss, these dynamics are especially important for considering the expression of the event of decolonisation70 in each area.

Colonial modernity, Wagner (2015:2) argues, inaugurates ‘a history of oppression while at the same time serving as a reference-point for resistance to oppression and struggle for justice’ (see also Chatterjee, 1986; Eisenstadt, 1999a:289). Among the varieties of modern experience, there is therefore a crucial difference between the European and the non-European:

The former generated modern commitments from problems they were facing internally; the latter experienced European expansion as external domination, and their modernity as commitment to individual and collective self-determination thus needs to work through the trauma of being long deprived of the possibility to self-govern their lives (Wagner, 2014:308- 309).

70 I make a distinction between the event of decolonisation (in Burundi and Rwanda, this is marked by formal political independence on the 1st July, 1962) and the process of decolonisation as an ongoing project (Fanon, 1965:30; Thomas et al, 2005:6). 168

In this chapter, I deal with the modern commitment to autonomy and its ambivalent relationship with colonialism, with reference to the specific empirical context of Rwanda and

Burundi in the period following the Second World War until the early 1960s. I am particularly concerned with the competing and conflictual interpretations among elites of the imaginary significations of modernity in Burundi and Rwanda in comparative perspective, and with how these interpretations are both enabled and constrained by the experience of modernity impinged by indirect colonial rule. I will also consider how such interpretations were shaped by precolonial legacies, and how they emerge in the global space opened up following the end of

World War Two. I call these interpretations, employing Peter Wagner’s (2010a, 2012) term,

‘societal self-understandings’.

MULTIPLICITY, ELITES, AND SOCIETAL SELF-UNDERSTANDINGS

My primary thesis in this chapter is that in the period under investigation, multiple interpretations of the modern social imaginary emerged and competed with each other in the form of programmes for independence. I suggest that we can see these movements in Burundi and Rwanda as a specific articulation of a transition from colonial modernity, which was the subject of the previous chapter, to what might be called a postcolonial modernity, via the

‘decolonial’ route. This lines up alongside theorisations of ‘successive’ modernities (Arnason,

2005; Wagner, 2010). Here, the plurality and multiplicity of modernity is not located solely in culturally-differentiated interpretations of select tenets of modernity which can be studied comparatively as spatially distinct units. Multiplicity is also identified temporally in transformations of modernity, defined globally and in the singular (Wagner, 2010:12, 2012:73).

Specific routes to and through modernity do not emerge in isolation but rather are entangled with the transformations occurring within a much more encompassing global modernity, transformations generated by the immanent critique and structuring activities of human collectivities (Wagner, 2012:32). 169

As Therborn has argued in terms that evoke the processual idea of successive modernities, colonial modernity was generally succeeded in all cases by ‘the road to modernity by anti-colonial rebellion’ (Therborn, 2003; 2011:60-61). As I suggested in chapter three71, my own preference is for the term ‘decolonial route’, rather than anti-colonial rebellion. In the cases of Burundi and Rwanda, the expression of this transformation took distinct forms;

Rwanda became independent as a republic via a highly ethnicised social revolution (Linden,

1977; Mamdani, 2001:103-131), Burundi as a constitutional monarchy via a national liberation movement (Chrétien, 1997:154; Nsanze 2003:36-60).

Elite groups were principal actors in the transformation of modernity in this sense

(Eisenstadt, 2000, 2005a; Therborn, 2003, 2010:74, 2011:61). It is with them that interpretations of an autonomous modernity emerged. As Therborn (2003:293) suggests, ‘often the tenor of elite social discourse provides the basis of general characterizations of a society as modern or pre-modern or traditional’. On the verge of independence, Chrétien (2003:287) argues, ‘the ‘social’ question in Burundi and Rwanda was less an agrarian one and more one of elites’. This group was an awkward agglomerate of ‘an educated micro-bourgeoisie… youth won over by individualism and the spirit of initiative, and rural folk attracted by non- agricultural occupations’, who found their aspirations ‘stifled under Belgium’s bureaucratic- clerical style of trusteeship’ and the neo-traditionalism that was the focus of the previous chapter.

Representatives of this elite group, the vast majority of whom were educated in Catholic seminaries, created political and social movements, and the tensions and conflicts between them reflected broader tensions over the interpretation of the modern social imaginary. In

Rwanda, the principle schism manifested between the monarchist party Union Nationale

71 See pages 78-79. 170

Rwandaise (UNAR) and the Bahutu Social Movement and the ensuing Parti du Mouvement de l'Emancipation Hutu (PARMEHUTU). The main vector of conflict here was undoubtedly ethnic and was centred on the institution of the monarchy. The overall triumph of

PARMEHUTU following the social revolution led to the establishment of a republic, led by

Grégoire Kayibanda, a seminary-educated intellectual and catechist who became the country’s first independence leader (Twagilimana, 2016:132).

In Burundi, the primary line of differentiation was between the Parti Démocratique

Chrétien (PDC) and the Union pour le Progrès National (UPRONA). The sorts of ethnic antagonisms that marked Rwanda entered Burundian politics at a later stage. The principal division on the verge of independence were between ganwa factions. Independent Burundi emerged as a constitutional monarchy under the leadership of the Bezi prince Louis Rwagasore, who studied at the Astrida catholic school and later in Belgium (Chrétien, 2003:310; Eggers,

2006:132).

That both Kayibanda and Rwagasore – among other key political party leaders, social activists and intellectuals – had been educated in missionary schools raises an important point about the significance of Christianity in the interpretations of modernity that are the themes of this chapter. As Delanty (2006:269) notes, ‘the emergence of modernity in Europe is to be principally attributed to Christianity, which has given to it its most distinctive and abiding characteristic as a condition that is defined by a time consciousness’. That elite social discourses oriented towards autonomous conceptions of modernity tended to come from seminarians is significant in this regard. Missionaries tended to be more explicitly occupied with a self-conception as carriers of modern ideas, and saw their civilising role in explicitly moral terms (Mamdani, 2001:113; Therborn, 2011:48). As we shall see, this had some paradoxical effects. For instance, in Rwanda, the Church legitimated both the purported 171

superiority of Tutsi elites and the ideas and principles underpinning the discourse of the Hutu social revolution.

The interpretations of autonomy, mastery, social justice, equality and so on clearly existed in tension with the solidifying legacy of indirect colonial rule. As such, to quote

Arnason (2003:342), we see how:

Analyses of colonial modernity must also deal with a closely related issue: the perpetuation of its distinctive pattern – in more or less modified form – after the end of colonial rule. But the transition from colonial to post-colonial conditions is indisputably a turning-point which opens up new paths for historical interpretation. An effective way of thinking through this, I suggest, is to draw on work presented in decolonial theory, namely the concept of ‘coloniality’ 72. Coloniality is best defined as ‘a situation of cultural, political, and economic domination that can be enforced in the absence of colonial administrations, which it has historically tended to outlive’ (Boatcă, 2014:368; see also Quijano, 2008; Mignolo, 2012:17; Ndlovu-Gatsheni, 2013:6). It is generally divided into two analytical components: the coloniality of power and the coloniality of knowledge.

The coloniality of power refers to the relations of economic dependency and appropriation first established with the emergence of the modern world-system, elucidated principally by Immanuel Wallerstein and developed by the Peruvian sociologist, Anibal

Quijano (2008:187):

The control of labor in the new model of global power was constituted thus, articulating all historical forms of labor control around the capitalist wage-labor relation. This articulation was

72 In various parts of this thesis, I have suggested that the decolonial theoretical conceptualisation of modernity, linked inextricably to colonialism in the neologism ‘modernity/coloniality’, is more totalising and one-sidedly negative than the concept of modernity I employ. To summarise, I have argued that such a perspective flattens a key agential/interpretive dimension of modernity (what Arnason identifies in the above quote as the conditions for ‘new paths for historical interpretation’), and that it also reiterates the modernity/tradition dichotomy presented in modernisation theory (albeit with the normative claims reversed). I am also aware of the specifically Latin American origins and subject matter of decolonial theory, and it is not entirely clear whether it is intended to apply or indeed is effective in application to Africa, Sabelo Ndlovu-Gatsheni’s (2013) work notwithstanding. Nevertheless, the concept of coloniality captures well those aspects which are, in Arnason’s above terms, ‘perpetuated… in more or less modified form – after the end of colonial rule’. 172

constitutively colonial. First, it was based on the assignment of all forms of unpaid labor to colonial races… Second, labor was controlled through the assignment of salaried labor to the colonising whites. The coloniality of labor control determined the geographic distribution of each one of the integrated forms of labor control in global capitalism. In other words, it determined the social geography of capitalism: capital, as a social formation for control of wage labor, was the axis around which all remaining forms of labor control, resources, and products were articulated. The coloniality of knowledge refers to the cognitive, representational and epistemic patterns that are intertwined with the material relationships outlined above. The coloniality of knowledge denotes a ‘complex process of deployment of global imperial technologies of subjectivation taking the form of translating and re-writing other cultures, other knowledges, and other ways of being, and presuming commensurability through Western rationality’

(Ndlovu-Gatsheni, 2013:33; see also Maldonada-Torres, 2007:242; Mignolo, 2012:22). As I argued in the last chapter, the key factor in the case of Burundi and Rwanda is the racialisation of history and tradition which marked colonial scholarship on the country and which was the basis for the reconstruction overseen by the Belgian administration.

The perpetuation of the distinctive pattern of colonial modernity, in the form of coloniality, imposes constraints on interpretations of modernity. However, I argue that the former does not totally curtail the latter. There is rather a tension between them. It is in this tension between imaginary significations of modernity (autonomy, mastery, equality, social justice, etc.) developed in the seminaries and disseminated in organs of an emerging public sphere, on the one hand, and the institutional and epistemic forms of coloniality on the other, that the concept of ‘societal self-understandings’, and their specific versions in Burundi and

Rwanda, ought to be situated.

‘Societal self-understandings’ is a term proposed by Peter Wagner, intended to move beyond the problems entailed in the concept of ‘society’ routinely employed in sociology and in more popular parlance: ‘rather than on high commonality among its members or on socio- structural cohesion, it focuses on communication between human beings about the basic rules 173

and resources they share, and on the sedimented results of such communication’ (Wagner,

2010a:56).

As I suggested in chapter three73, these basic rules and resources that are the concern of communication are conceptualised by Wagner as problematiques. These concern ‘a) as to what certain knowledge a societal self-understanding is seen to rest upon; b) as to how to determine and organize the rules for the life in common; and c) as to how to satisfy the basic material needs for societal reproduction’ (Wagner, 2010a:57; see also Wagner, 2008:4). They are labelled the epistemic, political and economic problematiques respectively. These are observable problems that any human society has to address. To speak specifically of ‘modern’ societal self-understandings is to recognise that ‘all these questions are truly open; that answers to them are not externally given but need to be found; and that, therefore, contestation of the validity of existing answers is always possible’ (Wagner, 2010a:57). Societal self- understandings are tied into Wagner’s (2008, 2009) theory of modernity ‘as experience and interpretation’ in the sense that they ‘refer to aspects and events that are specific to a given society, such as the moment of foundation… or a highly significant collective experience’

(Wagner, 2010a:57).

The specific interpretations and answers to each of the problematiques in Burundi and

Rwanda are limited by the persistence of coloniality. The distinct facets of coloniality impinge upon the capacity to address the problematiques in specific ways; the coloniality of power constrains the interpretation of and answer to the economic and political problematiques, and the coloniality of knowledge constrains the epistemic problematique. Colonialism, in this case, thus constitutes the experience upon which interpretations of modernity – and the resultant specific institutional forms – emerge.

73 Pages 70-72. 174

The idea of collective interpretation evoked here raises some questions that need to be addressed. The first concerns the notion of collectivity and the second concerns the interpretive capacities of human beings within such collectivities. In addressing these problems, I refer back to my previous discussion of philosophical hermeneutics74. I shall also draw on Wagner’s specifically cultural theory of modernity as discussed with Heidrun Friese (2000). They suggest that ‘to be able to underpin a theory of modernity, cultural analysis needs to be seen as interested in the ways human beings interpret the world and their being in the world’ (Friese and Wagner, 2000:34). I have suggested that human beings are ‘thrown’ into the world as beings who understand and interpret. Beyond this, by virtue human interdependencies, ‘being- in-the-world’ always entails ‘being-with-others’. As Friese and Wagner (2000:34) suggest:

Rather than being endowed with, or inheriting through socialisation, a set of cultural norms and values, human beings relate to the world and the interpretations that are offered about it, reflect upon the world and the interpretations and devise ways to deal with them. While the world and interpretations of it are always already there, they are not given, they need to be – selectively – appropriated and will be altered in the process… Those human beings are not atomistic monads and not individuals in the strict sense of the term; they are social beings. However, they are not unspecific parts of cultural collectives either, whose characteristic it merely is to be like other members of their culture and different from members of other cultures. Culture defined thus is conceived as fragmented and undetermined, in Arnason’s

(1993:10) terms, ‘an open-ended, understructured and imperfectly integrated complex of interpretive patterns, rather than as a value system or an ideological machine’. Inasmuch as culture concerns what binds individuals into collectivities, the creation of such a binding becomes ‘an interpretive act, the production of intersubjective meaning’ (Friese and Wagner,

2000:28). The production of intersubjective meaning includes specific narrations of historical events and the construction of forms of community through such narrations and the creation of symbolic bases for ethnic mobilisation, as well as institutional forms such as political parties, through which they can be expressed (Nagel, 1994). In summary:

74 Pages 89-101. 175

One needs to see the ‘modern’ commitment to autonomy and mastery as instituting an interpretative space that is to be specifically filled in each sociohistoric situation through struggles over the situationally appropriate meaning. Theoretically, at least, there is always a plurality and diversity of interpretations of this space, and never any unequivocally ‘modern’ way of arranging social life. This is where a theory of modernity links up to the hermeneutic and interpretive tradition of social thought. Any theory of modernity that aims at becoming more of a positive social science than a broadly hermeneutic position allows for will not only take a particular stance in social theory with all the implications known from the dispute on positivism and many other disputes before and after. It will in particular fail to understand the specificity of modernity, its inherent openness and self-avowed commitment to endless disputes over interpretation (Friese and Wagner, 2000:33-34 emphasis added). Therefore, those autonomous conceptions of modernity that emerged in Burundi and

Rwanda in the run-up to independence have to be seen as active hermeneutical engagements with the world. The imaginary significations of modernity were thus interpreted into the specific contexts of Burundi and Rwanda in the 1950s and 1960s, in the midst of rapid decolonisation and global turbulence. The subjects of this chapter were not, then, dupes or puppet-elites of western colonial administrations eager to retain an influence in the territories that they were departing. Rather, they were interpretive and creative agents possessing distinct orientations to the world, and future anticipations that had to be negotiated with the continued constraints imposed by the institutional order of colonial rule. In both Burundi and Rwanda, multiple modernities in competition and conflict terms were present in one societal setting

(Ichijo, 2013:30) and, owing to the solidifying experience of indirect colonial rule, the principle line of division was ethnic (though to varying degrees in each).

At this point, I have outlined the interpretive dimensions of modernity already discussed in chapter three and have introduced the notion of societal self-understandings. I have indicated that these societal self-understandings are multiple and conflictual, and are shaped in large part by precolonial legacies of differentiation and their transformation in the colonial entanglement.

I have said that any autonomous interpretations of modernity in the transition to independence had also to contend with coloniality, the persistence of the distinct economic, cultural, political and epistemic patterns and relationships of colonial rule following their formal disbanding.

Beyond this, however, the societal self-understandings under investigation here emerged in the 176

context of a specific situation that needs to be addressed. I contend that the transformation of modernity operative at the global level, of which decolonisation in the broad sense was both emblematic and productive, opened up a space for interpretations to emerge. As such, it is now necessary to turn to this transformative global context.

SITUATING BURUNDI AND RWANDA IN TRANSFORMATIONS OF THE GEO- HISTORICAL ENTANGLEMENTS OF MODERNITY

As Fanon (1965:55) stated, ‘a colonised people is not alone. In spite of all that colonialism can do, its frontiers remain open to new ideas and echoes from the world outside’.

Events in Burundi and Rwanda from the mid-1940s to the early 1960s, culminating in their independence, cannot be seen in a vacuum and as such ought to be placed in a broader context and frame of reference. At the regional and continental levels, events in neighbouring areas like the Congo, Tanzania and Uganda had an undeniable impact in terms of the spread of nationalist sentiment, and there was considerable cross-national communication between

évolué figures, such as Grégoire Kayibanda in Rwanda and Louis Rwagasore in Burundi, and figures like Patrice Lumumba and Julius Nyerere (Lemarchand, 1970:84; Chrétien, 2003:302-

303). Beyond these regional entanglements, however, the independence movements in Burundi and Rwanda ought to be situated more broadly in the context of the transformation of global modernity. This involves a shift, from what Wagner (2012:163-167) – building on the above outline of the theory of successive modernities – calls a global organised modernity enduring from the 1880s to the 1960s, to the destructuring of organised modernity attributable to processes of globalisation beginning in the 1970s.

Such a shift prompts us to consider a range of new external historical actors and structures, each of which exercised considerable influence on trajectories towards independence. The newly-created United Nations and the intensification of Cold War animosities, in addition to changing attitudes in the Catholic Church – carried into Burundi and 177

Rwanda by a new generation of missionaries – and the Belgian colonial administration, would all play significant roles in the fates of the two countries (Shaw, 2002:88; Steinmetz, 2014:80).

Although press reportage in the 1990s might have given the contrary impression, the entangled histories of post-independence ethnic strife in Burundi and Rwanda were not at all disconnected from these global transformations.

The transformation of modernity that Wagner (2010, 2012:163-167) has conceived of as the ‘deconstruction of global organized modernity’ is well expressed in the post-war crisis of empire. Numerous reasons have been supplied for this crisis, including the rise of western- educated elites who turned this education against the colonisers in programmes of national liberation, as well as the post-World War Two downturn in European economies, which exacerbated the strains of the Great Depression.

The political disarray that had been produced during the war also significantly weakened the metropolitan imperial administrations. The Belgian government, for instance, had been forced to move to London because of Nazi imperialism (Thomas et al, 2008:415-416;

Young, 2008:702). Furthermore, the experience of war as felt by the Belgian colonial subjects intensified anti-colonial sentiments (Young, 2008:702). During this period of exile in London, the colonial administration relied on funds from the colonies, and the extractive and exploitative labour of the population was intensified (Young, 2015:108-109). In Ruanda-

Urundi, which felt the war effort to a lesser degree than Belgian Congo, this was compounded by famine in 1943-4 due to poor rainfall (Young, 2008:704). The post-war intensification of rivalries between east and west led to competition for African and Asian loyalties, and there was also evident an increasing reticence concerning empire among metropolitan publics and governments (Steinmetz, 2014:85). 178

The United Nations emerged out of the ashes of World War Two and the emergence of the international trusteeship system. Its basic objective was ‘to promote the political, economic, social and educational advancement of the inhabitants of the trust territories, and their development towards self-government and independence…’ (United Nations, 1945:15). This was the catalyst for a number of reforms in Ruanda-Urundi (Duarte, 1995). However, in this case, there remained a tension between the aim of self-government, ostensibly emerging from the humanitarian and egalitarian ethos of the United Nations, and the specificity of Belgian rule, particularly as regards its paternalism75.

A number of Visiting Missions of UN representatives occurred from the late 1940s. In

1948, the first of them praised the level of economic and social welfare provided by the

Ruanda-Urundi administration but decried the lack of provision for political advance, as did another mission in 1951. A decree was issued in 1952 that proposed a hierarchy of councils in order to provide for limited African participation. Its impact was minimal, however, owing to the distant and indirect nature of UN consultations, and in Rwanda especially the decree was filtered through the chiefly hierarchy (almost all Tutsi) (Young, 2008:706-707). As we will see, the relative ineffectiveness of UN democratisation processes and the exclusionary consequences for Hutu people was a catalyst for revolutionary sentiment in Rwanda. It was only later on, from 1959 to 1962, that a period of accelerated democratisation was inaugurated, which led in 1960 and 1961 to the establishment of popularly elected institutions of government both at the local and central levels, and ultimately to the independence of each territory as a separate political entity (Lemarchand, 1970:80-81).

75 An expression of this paternalism is presented in the words of a Belgian special representative to the Trusteeship Council, who said this of Belgian policy: ‘The real work is to change the African in his essence, to transform his soul, [and] to do that one must love him and enjoy having daily contact with him. He must be cured of his thoughtlessness, he must accustom himself to living in society, he must overcome his inertia’ (in Duarte, 1995:276). 179

Despite their small size, Burundi and Rwanda were also conceived of as strategic countries in the rivalry between the east and west. For Mao Zedong, Burundi was ‘the stepping stone for reaching the Congo’ (in Lemarchand, 1970:391). The CIA established a base in of operations in Rwanda following its independence (Chrétien, 2003:313). Colonial policy on the eve of decolonisation was often couched in terms of the scission between the capitalist west and the communist east. In Burundi, UPRONA were discouraged because of their communist leanings and relationships with figures like Lumumba and Nyerere, and in Rwanda the Hutu revolution was supported by the colonisers on the basis that the explicitly anti-Western Tutsi elite associated with the Nyiginya court were professed to have communist sympathies

(Lemarchand, 1970:175-176).

The Belgian colonial administration also became embroiled in the emerging crisis in

Congo, itself a key theatre for east-west conflicts, and they diverted a significant number of resources from Ruanda-Urundi to try to bring order (Vanthemsche, 2012:89-98). We have already seen how Belgian Congo was deemed of more importance than Ruanda-Urundi, and that the latter’s operation tended to be oriented to serving the needs of the former (Gahama,

2001). Accordingly, the experience of decolonisation in Congo had a large follow-on effect in

Ruanda-Urundi (Latham-Koenig, 1962).

Perhaps the most noticeable characteristic of the independence of Congo was its rapidity. Less than a decade before formal independence, it had been seen as a very distant prospect. The paternalistic style of Belgian administration resulted in little provision being made for the realisation of independence. Even Patrice Lumumba, who spearheaded the anti- colonial movement, admitted in 1955, only five years prior to independence, that Congo still needed to reach ‘the more advanced degree of civilisation and the required political maturity’ to advance independently (in Young, 2015:109). As we will see below in the case of Ruanda-

Urundi, fissures appeared in the territory when a group of Catholic missionaries declared 180

support for Congolese emancipation. French-Flemish tensions also spilled into Congo. The

Leopoldville riots of 1959, spontaneous and leaderless, spurred anticolonial sentiments still further. This, of course, was set against the background of the accelerating development of decolonisation across the African continent (Thomas et al, 2008:388; Young, 2015:110-115).

Also important is a transformation in the allegiances of the Catholic Church, particularly in Rwanda. Following the war, the Catholic Church in Rwanda switched allegiance to Hutu and developed a strong concern with social equality as distinct from earlier conceptions of racial inequality. This was also true of the Belgian administration, which had hitherto been a vector of Tutsi-favouring racial stratification. This change was tied to the arrival of a new generation of missionaries in Ruanda-Urundi. In contrast to earlier missionaries, they were from relatively humble backgrounds and identified with the ‘downtrodden’ Hutu. In part, this was to do with their own experiences of disenfranchisement in Belgium, marked as it was by sharp divisions (Lemarchand, 1970:107; Linden, 1977:222; Mamdani, 2001:113). It is also conceivable, I suggest, to see this as a response to the ‘trauma drama’ of World War Two and the Holocaust in Europe, a response which gave rise to a tragic paradox: the memory of national socialism among these missionaries, and their associations of totalitarianism with the Nyiginya kingdom, had unintended effects that are noticeable in Hutu ethno-republican discourse during bouts of genocidal violence in the postcolonial period (see Alexander, 2016:9).

In Burundi, by contrast, the Belgian administration tried and failed to stem the tide of nationalist sentiments that was crystallising around the charismatic figure of Louis Rwagasore and the UPRONA party. Governor J.P. Harroy said: ‘We won the Rwandan elections, but we lost them in Burundi’ (In Chrétien, 2003:312). As in the previous chapter, we ought to avoid associating discourses of European racial superiority and the civilising mission of colonialism straightforwardly with the implementation of clearly defined colonial policies. In public relations the formal decolonisation of the Belgian colonies was described by the colonisers in 181

humanitarian terms that implied that they themselves held the key to the freedom of the subjects of their civilising mission, but in reality it was chaotic and unplanned, with some destructive unintended consequences.

Such context should not be overemphasised as a determining factor in independence movements. It would be inaccurate to see independence as the result of the concatenation of global forces, or even worse the result of colonial powers granting autonomy benignly after seeing the error of their ways. This quite clearly denies the colonised populations any sort of agential role in the event of independence. However, equally misleading is the idea that anti- colonial resistance was constituted by some effervescent homogeneity on the part of a singular

‘colonised’ population (Thomas et al, 2008:5). It is rather the case, as Wagner (2015:7-8) puts it, that ‘the ‘entanglement’ of forms of modernity should be seen less as occurring between constituted societies than, rather, through the interpretative agency of human beings’. What I explore below, and will be identified as having significant consequences for postcolonial developments, is that in terms of the societal self-understandings that emerged during independence movements in Burundi and Rwanda, there was much differentiation at the level of human agency.

DISENTANGLING TRAJECTORIES TOWARDS INDEPENDENCE

RWANDA

The form that the formal process of independence took in Rwanda between the years

1959 and 1961 has frequently been characterised as a social revolution, both in postcolonial governmental and popular discourse and in Rwandan scholarship (Lemarchand, 1970; Des

Forges, 1999:36; Mamdani, 2001:103; Chrétien, 2003:299). The distinctive feature of this social revolution is that the principal vector of stratification and grievance was ethnic; it was a

Hutu revolution. As we shall see in the next chapter, the revolution was a foundational event 182

in the narrative constitution of the postcolonial state (Chrétien, 2003:306), and was intimately connected to the ideologisation and sanctification of violence in the postcolonial period

(Eisenstadt, 1996a, 2005a). Fear of its reversal was mobilised by successive post-independence governments to incite violence against Tutsi people, defined in homogenous terms as collectively responsible for insurgencies led by Tutsi rebels from south-western Uganda who themselves had been displaced by the revolution (Des Forges, 1995; C. Newbury, 1998).

The corporealisation of Tutsi and their purported collective responsibility for the country’s troubles was nonetheless a post-independence phenomenon. Though there was undoubtedly a violently anti-Tutsi component to the revolution, this was insofar as Tutsi people were emblematic of the central court and the injustices perpetrated under colonial rule. The principal target for the Jacobin tenors of the revolution was the monarchy. It was only later that an anti-Tutsi sentiment extended beyond the court, chiefs and sub-chiefs, and manifested in a logic of collective culpability (Lemarchand, 1970:125; Newbury and Newbury, 2000:869).

In short, two different kinds of anti-monarchist orientations emerged in different localities at different points in time, and converged in the same direction (Lemarchand,

1970:99). First, there was the restorative movement that came into being in the north-western

Bakiga region of the country, a movement connected to the Nyabingi cult, a millenarian protest movement which emerged in response to precolonial Nyiginya expansion and the centralisation processes that accompanied colonial rule (Cohen, 1989:292; Rutanga, 1991). Second, there was a social reform movement that developed principally in southern and central Rwanda in the 1950s, whose activities are expressed most clearly in the publication of the Bahutu

Manifesto of 1957 and the emergence of PARMEHUTU (Lemarchand, 1970:103; Mamdani, 183

2001:107-108). A monarchist party, UNAR, emerged in response to these movements and was largely perceived as defending Tutsi power interests. I shall focus on each of these below76.

Theda Skocpol (1979:4), in her classic comparative study on the subject, characterised social revolutions as ‘rapid, basic transformations of a society’s state and class structures… accompanied and in part carried through by class-based revolts from below’. The unique factor of social revolution is that, in contrast to rebellion or political revolution, ‘basic changes in social structure and in political structure occur together in a mutually reinforcing fashion. And these changes occur through intense socio-political conflicts in which class struggles play a key role’.

In many ways, the Rwandan revolution constituted a social revolution in this sense, though with some peculiar characteristics. Owing to the reconstruction of tradition that had occurred during the colonial encounter, class categories were heavily imprinted with a racial logic (Kuper, 1971, C .Newbury, 1998). Thus, soon after independence, the anthropologist

Helen Codere, who had been present in the country in the years 1959-1960, suggested that the events had led to ‘irreversible shifts for both the Tutsi and the Hutu in fundamental attitudes, in social relationships and in access to economic benefits and to political power’ (Codere,

1962:63). There were fatal attacks on Tutsi collines and mass displacements, a dismissal of well over half of Tutsi customary authorities and their replacement by Hutu burgomasters, and

76 Several moderately and explicitly class-based parties were also present, who attempted to transcend ethnic divides and appeal to ‘common people’. These included the Association pour la Promotion Sociale de la Masse (APROSOMA) and Rassemblement Démocratique Ruandais (RADER). A major characteristic of post-independence politics in Rwanda, seen in the 1950s and especially in the genocidal context of the 1990s, is the hollowing out of the moderate middle ground. We should not forget, in Catherine Newbury’s (1998:16) words, the ‘courageous voices of moderation that called for inclusiveness, the restructuring of power relations, and the necessity to attend to the needs of all Rwandans, regardless of ethnic background’. In the 1994 genocide in Rwanda, as we shall see, moderate politicians and activists were among the first to be killed. Here, in the context of the 1950s, they were marginalised by other means (see also Des Forges, 1999:36). 184

an abolition of the monarchy, all presided over by the political and military structure of the

Belgian administration (Chrétien and Dupaquier, 2007:18).

In order to explain these transformative moments, Skocpol argued, it is necessary to

‘find problematic, first, the emergence (not ‘making’) of a revolutionary situation within an old regime’ (Skocpol, 1979:18). As such, the revolutionary situation in Rwanda ought to be understood as having emerged out of the contradictions of indirect colonial rule, and the ‘dual colonialism’ of the expansionist Nyiginya court and the Belgian administration (C. Newbury,

1988; 1998:12) against the background of the broader transformations outlined above. Due to the changing attitudes of Catholic missionaries during the 1950s, for instance, more and more

Hutu people were permitted into education beyond primary level, and in turn they formed part of a disaffected group whose expectations and aspirations were disappointed by the institutional favouritism towards Tutsi people (Longman, 2001:169). This more egalitarian ethos was also expressed in the abolition of ubuhake clientship in 1954 (Lemarchand,

1970:127; Linden 1977:226-227).

Perhaps the most emblematic of the new missionaries was Monsignor Perraudin, who was emblematic of an increasing post-war identification of colonial administrators and missionaries with the Hutu cause. As I have suggested, this was influenced by the nascent ethnic conflicts in Belgium, as well as fears about the appeal of communism and atheism to the majority of Rwanda’s population (Carney, 2011:88). It was also facilitated by the legacy of the

Second World War and genocide in Europe. As Perraudin said of the UNAR party and their purported defence of monarchist and ethnic interests: ‘the UNAR party seems to wish to monopolise patriotism and to say that those who are not with them are against the country. This tendency strongly resembles the national socialism that other countries have known and which has done them so much harm’ (in Linden, 1977:266; see also Lemarchand, 1970:161). It was also through the connections with the Church that figures like Grégoire Kayibanda were able 185

to publish in daily newspapers like Temps Nouveaux d’Afrique, which commanded a readership in Ruanda-Urundi among literate Africans and Europeans alike (Lemarchand, 1970:151). The journal Kinyamateka, of which Kayibanda was the chief editor, was a vehicle used to highlight the social problems faced by Hutu people (Rothermund, 2006:165).

In the 1950s, a Hutu movement developed in southern and central Rwanda. Its demands were initially reformist rather than revolutionary, couched in what was a highly ethnicised language drawing on egalitarian-democratic ideas (Lemarchand, 1970:112). It is expressed most clearly in the publication of the Bahutu Manifesto of 1957 (United Nations, 1957:39-42; see also Lemarchand, 1970:103; Mamdani, 2001:107-108). Composed by Hutu intellectuals and former seminarians, the document presented a fundamental challenge to what they referred to as the ‘feudal system’ of monarchical Rwanda, and called for ‘the economic and political emancipation of the Muhutu from his traditional subjugation to the Hamites’ (United Nations,

1957:41).

The presentation of their relationship to the Belgian administration was ambivalent. On the one hand, there were warnings that ‘the departure of the Europeans might plunge [Hutu] into worse slavery than before’ and that Hutu would express ‘the right to refuse to co-operate in the efforts to attain independence’ unless the structures of Tutsi domination were overturned

(United Nations, 1957:39; see also Young, 2008:709). On the other, there was expressed an historical experience of a ‘colonialism in two stages: the Muhutu must bear with the Hamite and his domination and with the European and his laws’ (United Nations, 1957:39).

Beyond this, there were relatively modest demands, such as for legal recognition of individual land ownership ‘in the Western sense’, the abolition of corvée labor, the promotion of a rural credit fund, greater ‘economic union between Belgian Africa and the metropolis’, freedom of expression, the promotion of Hutu people to positions of public office, and so on 186

(United Nations, 1957:40-41). Among those who signed the manifesto was Grégoire

Kayibanda, who in the same year established Mouvement Social Muhutu along with other

Astrida graduates such as Joseph Gitera and Balthazar Bicamumpaka (Twagilimana, 2016:132).

This movement took political form in the Parti de Mouvement de l’Emancipation Hutu

(PARMEHUTU), also known as Mouvement Démocratique Républicain. In their own words, the movement was ‘charged with the social, economic and political liberation of the population’, the ‘democratisation of the institutions of the country’, and the ‘definitive abolition of the feudality that has handicapped the general progress of the nation’, which would be achieved through the ‘total abolition of the triple myth of the Tutsi feudal colonialists,

‘Kalinga[drum]-Abiru[traditional court historians]-Mwami [the king]’ (PARMEHUTU, 1960; see also Chrétien, 2003:301).

In 1959, responding to these Hutu movements, UNAR emerged. It was ostensibly dedicated to ‘the union of all Rwandese for the purpose of achieving true progress in all spheres’

(in Lemarchand, 1970:158). The theme of kingship found its way into their numerous publications and pamphlets, and it was often suggested that Hutu, Tutsi and Twa people had resided for aeons in harmony with each other before successive European administrations drove a wedge between them with their racial discourses. This fermented an attraction to anti-

Western communism, albeit one never divested of a reverence for tradition. As put by Paul

Rubeka (in Lemarchand, 1970:159), the party’s leader, the aims of UNAR were stipulated along the following lines:

The whole of Africa… is struggling against colonialism, the same colonialism which has exploited our country and destroyed our ancestral customs in order to impose alien ones upon us. The goal of our party is to restore these customs, to shake off the yoke of Belgian colonialism, to reconquer Rwanda’s independence. To remake our country we need a single party, like UNAR, based upon tradition and no other ideology. He who does not belong to this party will be regarded as the people’s enemy, the Mwami’s enemy, Rwanda’s enemy. 187

Behind the rhetoric of anti-colonial national unity, the fidelity to the tradition, and the refusal to acknowledge any tension between Tutsi and Hutu people77, it was possible to detect defences of parochial Tutsi interests in the discourse of UNAR. In the Bahutu Manifesto, a reference was made to ‘the thought, the pet method of the Administration in our country, which is sarcastically called ‘Umuco w’Igihugu’, ‘respect for the culture and customs of the country’’

(United Nations, 1957:40). Occasionally such brazen ethnicist interests burst out into the open.

In response to the Bahutu Manifesto, for instance, twelve ‘grand clients’ of the court wrote that:

Those who demand the joint division of property are those that have between them bonds of brotherhood. But relations between us (Tutsi) and they (Hutu) were always until now based on serfdom; so there is between us and them no foundation of fraternity. Indeed what relationships exist between Tutsi, Hutu and Batwa?… Since our kings conquered the country, killed their petty kings and thus enslaved the Hutu, how now they can they claim to be our brothers? (Nkundabagenzi, 1958:35-36). What gave this elite a sense of solidarity, in addition to the emergence of the threats posed by a Hutu counter-elite, was a shared sense of occupying a superior social stratum legitimated by history. This was particularly supported by members of the Tutsi clergy who were educated to a higher degree than the chiefs and as such were able to compare Rwandan civilisation to others in order to ascertain its specificity (Lemarchand, 1970:136-137). Chief among these was Alexis Kagame, one of Rwanda’s most brilliant and well-known historians and promulgators of ethnophilosophy, particularly Bantu philosophy 78 . Kagame was particularly sceptical of the emergence of ‘certain egalitarian tendencies’ among politically conscious Hutu people. For Kagame, and others around him:

The path of progress cannot stray away from our traditional heritage… regardless of the type of socio-political system adhered to, one must avoid humiliating traditional authorities, either by disregarding their claims to leadership or casting discredit upon them in front of their subjects under the pretext that everybody is equal. The conclusion the masses are likely to draw

77 A 1957 ‘Statement of Views’ (UN, 1959) made by the High Council of Ruanda (all Tutsi) made reference to the ‘racial problem’ in the country only insofar as it involved the potential conflict between white European residents and black Africans. There was no mention of any of the contents or demands made in the Bahutu Manifesto of the same year. 78 See earlier discussion on pages 115-116. 188

from all this is that progress, freedom, in short everything, implies contempt for traditional authorities’ (Kagame, 1945:56). The reformist orientation of the Bahutu Manifesto was over time supplanted by more revolutionary sentiments that developed in response to the failure of moderate Hutu politicians to bring about the social and political reforms contained therein (Lemarchand, 1970:114). In

November 1959, a peasant jacquerie occurred. This was directed against the authority of the chiefs, rather than against the legitimacy of monarchical rule per se. It was only in the northern region, among the Bakiga people, that a more coordinated and instrumental goal could be detected. Here, the movement was intended to replace (as well as displace) the chiefs in office with descendants of traditional Kiga lineages. According to Lemarchand (1970:114-115), ‘here we find in germ many of the ideas which later found expression among the northern Kiga elites: a conspicuous hatred of Tutsi chiefs, a tendency to revive an idealised past, and a marked aversion towards all forms of centralised authority’.

Such a perspective reveals an orientation different to the more reformist tones of the

Bahutu Manifesto and the revolutionary turn that took hold among southern and central Hutu after the failure to meet many of the manifesto’s aims. Nevertheless, the movements in the south and centre of the country converged with those in the north. The northern leaders of the revolution were representative of Bakiga lineages native to northern Rwanda (Lemarchand,

1970:103). Kiga refers to an area in present-day northern Rwanda and southwestern Uganda.

In the precolonial era, it was known as the Ndora kingdom and settled by people known as the

Bakiga – ‘people of the mountains’ – who demonstrated a linguistic similarity with the

Rwandan state, but none in terms of political and social institutions (Mamdani, 2001:52).

Under Rwabugiri’s expansionist drive in the nineteenth century, the region had managed to resist much incursion by the precolonial state and was only tenuously incorporated by the time of the colonial entanglement. It was thus a particular area of focus under the further expansion 189

of the court that occurred with the collaboration of successive colonial administrations

(Newbury, 1988; Prunier, 1995:19; Jefremovas, 1997:95, 100; Mamdani, 2001:70).

It is in this context that the Nyibingi resistance movement ought to be situated. The movement, a messianic possession cult, challenged colonial occupation and administration in the northern areas of Rwanda, as well as in parts of Congo and in the Kigezi district of Uganda, and played an important integrative role (Ranger, 1968:451; Chrétien, 1972; Mamdani,

2001:72). However, it built on an already existing resistance to the incursions of the Rwandan state. Colonialism here was most sharply experienced as a ‘dual colonialism’, comprised of

Tutsi and European colonisers (Newbury, 1988; Mamdani, 2001:71). Hutu and Tutsi did not exist as identity terms until this period and there existed a strong sense of regional autonomy, a distinction recognised by colonialists. As put by a commentator who worked at Astrida, the

Bakiga were a ‘simple mountain people… with unpolished manners and energetic habits’

(Leurquin, 1963:46). Thus, there emerged a distinct orientation among northern leaders of the revolution. As Lemarchand (1970:103) puts it:

Although Western acculturative influences undoubtedly played a part in fostering new expectations about the future, what was involved here was a desire not so much to bring the existing political system in line with egalitarian norms as to adapt it to pre-Tutsi norms, to standards of behaviour patterned along an indigenous ‘clanic’ tradition. More was contemplated than a mere reversal of Tutsi-enforced statuses; the aim was nothing less than a rehabilitation of traditional, pre-Tutsi statuses. Following the jacquerie, a more sustained period of revolutionary social and political upheaval was inaugurated, which took place between the months of November 1959 and

January 1961. The aim was no longer to change the personnel of governing administration, but rather to bring about a more fundamental social transformation. It was during this period that ethnic violence increased and became more instrumentalised, becoming in Lemarchand’s

(1970:115) words ‘one of the prime accelerators of social change. Acts of terrorism against

Tutsi elements became so numerous that the administration could no longer repress them 190

(assuming it ever had the inclination to do so seriously and effectively); in turn, Tutsi-instigated counter-terrorism picked up momentum, leading to a society where ethnic antagonisms penetrated the whole of society’.

During this time, we can see an evolution of the target of violence, which moves from chiefs to the monarchy and then finally to the Tutsi community as a whole, as a greater sense of ethnic corporeality emerged (C. Newbury, 1998:13). The result was the systematic exclusion of Tutsi people from political life, mass killing of Tutsi people reaching genocidal proportions, and the expulsion of many thousands of Tutsi people into neighbouring areas.

In the emergence of democratisation in Rwanda, to borrow Michael Mann’s (2004) terms, the demos became equated with the Hutu ethnos. This ethnonationalism was explicitly supported by the Belgian administration. Colonel Guy Logiest, an admirer of apartheid named special resident during the Congo crisis, oversaw an effort to ‘de-UNARmize’ Rwanda in ‘an operation that involved dismissing half the chiefs and three hundred out of five hundred subchiefs and replacing them with Hutu chiefs and burgomasters’ (Chrétien, 2003:304). Codere

(1962:64, see also D. Newbury, 2005:272) notes how the most common ‘revolutionary act’ was to burn down Tutsi houses in a particular neighbourhood, ‘after due warning so the occupants could leave in advance, and usually without pillage, or any act of violence against persons’. And although corporeal views of ethnicity were projected in political discourse, political action diverged in significant ways. There remained a class component: ‘In the conflicts associated with the Rwandan Revolution of 1959-1961, rural dwellers often distinguished powerful and wealthy Tutsi from Tutsi commoners. The primary targets of attack were those who were in a position of power’ (C. Newbury, 1998:13).

The new independent order of Rwanda was consecrated in January 1961, in Gitarama.

The monarchy was abolished and Rwanda became independent on 1st July 1962 (Lemarchand, 191

1970:115; Kuper, 1970:372). As we shall see in the next chapter, internal order in post- independence Rwanda was secured at multiple points on the existence and ramping up of an external threat posed by those Tutsi exiles created by the revolution. There is an intrinsic relationship of this order to violence, against an ‘internal enemy’ who became more and more indistinguishable from the ‘external enemy’.

BURUNDI

In Burundi, the emergence of the liberatory situation out of the crisis of the old regime took a different form. As it was about to recover its independence as a monarchical state, the factionalism of ganwa politics became reinvigorated. There also emerged a wide range of other political factions, some of which had revolutionary aspirations. Twenty-three political parties were officially registered by 1961. Of these, the Parti de l’Unité et du Progrès National

(UPRONA) and the Parti Démocratique Chrétien (PDC), were the most significant. Their leaderships reflected the ganwa rivalries, the Bezi lineage being associated with UPRONA and the Batare with the PDC, as well as antagonisms between those educated at Astrida and those at the seminary in Mugera (Lemarchand, 1970:324; Chrétien, 1997:154; Daley, 2006:666;

Chrétien and Dupaquier, 2007:19; Young, 2008:715). Nationalist sentiments and developments, then, channelled dynastic feuds that were precolonial in origin.

Reasons for variation in modern societal self-understandings can be located in differences among Bezi and Batare princely factions and their experiences of colonial rule. A number of Batare chiefs earned respect and favour early on in the colonial administration. They also embraced symbols and lifestyles of European modernity during colonial rule; thus, in the nationalist vocabulary of the 1950s, the Batare came to be identified as moderate collaborators with the Belgian administration, while the Bezi were defined as radical anti-colonialists

(Lemarchand, 1970:336; Young, 2008:715). There was a more straightforwardly anti-colonial 192

nationalist sentiment in Burundi. This cut across party and identarian lines, initially, and was manifested in a spirit of national unity explicitly opposed to the divisionist ethno-histories of the colonisers (Chrétien, 1997:155).

During the late 1950s and early 1960s, these nationalist movements were ‘modernised’ in the sense that they incorporated membership of the population at large, introduced progressivism into the ideological programmes, and adapted their strategies to a system of government based on elections (Lemarchand, 1970:325; Nsanze, 2003:33-61). In Burundi, by contrast to Rwanda, the Belgian administration tried and failed to stem the tide of nationalist sentiment of the 1950s.

This nationalism crystallised in particular around the charismatic figure of Louis

Rwagasore, the figurehead of UPRONA party. Rwagasore was of the Bezi lineage, had studied in Belgium, and was a Tutsi who had married a Hutu. He was also the son of Mwami

Mwambutsa IV, and thus ‘he was able to be the acknowledged representative of the Bezi but stand outside of the Bezi-Batare conflict as a true national figure, while still having obvious ties to the still popular crown’ (Eggers, 2006:132-133). Rwagasore personified the growing attachment of the populace to the symbols of the crown, in the context of the harsh exactions of chiefs and ganwa figures who were the indirect agents of Belgian colonialism (Weinstein,

1965:12; Lemarchand, 1970:294). This nationalist sentiment of attachment can be seen in the motto of UPRONA – ‘One God, one Mwami, one Country’ (Lemarchand, 1970:294).

UPRONA under Rwagasore was able to combine the populist persuasions of the rural masses and their grievances about the abuses of the chiefs, subchiefs and the colonial administration, with the widespread attachment to the symbols of kingship (Lemarchand, 1970:329; Nsanze,

2003:39). Strong attempts were also made to divide political positions among Tutsi and Hutu figures. UPRONA also appealed broadly to workers, particularly through Rwagasore’s co- 193

operative movement, and realised that it had to placate and work with chiefs (Lemarchand,

1970:330-332).

UPRONA was able to incorporate elements of tradition into modern societal self- understandings to a much greater degree than the UNAR monarchist party in Rwanda, precisely because the role of the court had been marginalised in the experience of colonialism.

Rwagasore’s commitment to the monarchy was conditional ‘only insofar as the regime and its dynasty favoured the genuine emancipation of the Murundi people’ (in Lemarchand, 1970:330).

As stated in the UPRONA (in Eggers, 2006:158) party’s second manifesto:

The Burundi monarchy is constitutional, and wishes to see the constitution of the realm adapted to a modern state… UPRONA favours the democratisation of institutions… and will firmly and tenaciously combat all forms of social injustice, regardless of the system from which they may come: feudalism, colonialism or communism… UPRONA favours the election of the chiefs and subchiefs by the population, and will combat with all its forces those who seek to destroy the unity of the country. It is clear that far from advocating a return to a feudal era of princely rule, or a complete break with the past, UPRONA strove to modernise the kingdom by using the symbols of kingship as a means of absorbing and challenging change.

It was this aspect of UPRONA’s orientation that a community organiser from rural

Burundi called Paul Mirerekano tried to make explicit in a very widely-read pamphlet called

Mbwire gito canje (Listen my Son), published in 1961 (Lemarchand, 1970:330; Nsanze,

2003:39). To quote from a passage (in Lemarchand, 1970:306-307):

Let the Owner of the Drum, the Mwami of Burundi, reign. Let mwami-ship be strong, let it strengthen public order and the union of all Barundi in peace and justice!… Remember the beautiful customs of the realm… Too many people have forgotten the customs and traditions of the past, and therefore no longer know the proper behaviour to adopt towards others… One no longer makes a distinction between an eminent guest, close relatives, a mere friend and ordinary people… Solicit each other’s help, respect each other, forgive each other… The Barundi who are mindful of the established order know that to solicit each other’s assistance is the basis of love and unity among men. The attachment to cultural revivalism was an expression of anti-colonialism. Beyond the material disenfranchisement of the rural population and grievances towards the chiefs and 194

the colonial administration, there were serious attempts to excavate Burundian cultural traditions and adapt them to the demands of autonomous postcolonial modernity. This involved a return to oral sources among historians and the broadening of horizons to situate the country in the geographical context of east and central Africa, as against the racial histories of successive invasions in the Hamitic Hypothesis. Relatedly, there was a methodical questioning of notions of racial-ethnic separation between groups in a society of profound cultural homogeneity, the precise study of the nature of the ancient political system of Burundi in all its complexity, and the analysis of colonial rule and its effects on society, especially migration towards British central Africa and Congo (Chrétien, 1997:154-156). These resources provided routes to modern democratic construction and answers to the epistemic and political problematiques outlined by Wagner.

The PDC, led by Joseph Birolo of the Batare lineage, partook in this reconstruction too, even if it was largely perceived as more collaborative with the colonial administration than

UPRONA. Subsequently, its support was largely confined to those connected to the Batare lineage, and also some Burundians who had been educated at Astrida (REF). Abundant coverage was given in the PDC programme to questions such as ‘the problem of youth’, ‘social security’, ‘social relations’ and ‘public assistance’, all of which UPRONA tended to deemphasise in favour of the populist issues of national unity, monarchical legitimacy and independence (Lemarchand, 1970:335). The PDC was nevertheless more ambivalent about the virtues of self-rule than was UPRONA with its exhortations to ‘immediate independence’. ‘The most important thing we can say about independence’, according to a PDC spokesman welcoming the UN visiting mission:

is that it should not be hurled at us too quickly, without preparation… Independence can only be conceived as a long-run objective, not to be attained before the implementation of a double plan of progressive economic and political emancipation. For, after all, we are a poor country, and while we do have a political structure of our own, by no stretch of the imagination can it be said that it has prepared us for democratic government (in Lemarchand, 1970:336). 195

UPRONA was more popular, and upon independence eventually prevailed as the leading party, but it possessed a number of institutional shortcomings. Firstly, there was the overtly personalised character of UPRONA under Rwagasore’s charismatic authority, which manifested itself in the mirroring of the party’s fortunes with those of its leader. Second was the social context in which the party operated and the concomitant difficulty of reaching people outside the capital, Bujumbura. The mountainous physical environment of the country and the lack of transport infrastructure presented large difficulties in facilitating effective political mobilisation, and furthermore the legacies of the precolonial social system made political recruitment difficult on account of the proliferation of relationships based on client age and personal loyalty (Lemarchand, 1970:333). For Augustin Nsanze (2003:40), the nationalism of

UPRONA was founded on a rejection of colonial power and the aim of retaining as much sovereignty as possible, but there was also an element of the party that aimed to protect the interests of the ganwa princes. They were severely inhibited by the disciplinary actions of the

Belgian trust authorities, who were extremely anxious about the relationships between

UPRONA figures and leftist nationalist figures in newly independent Congo (Shyaka, 2008:6).

Louis Rwagasore was assassinated in 1961 by a Greek mercenary, not long before independence. The role of Belgium in this event is highly contested, but it was certainly the outcome of factionalism between Bezi and Batare factions. Rwagasore’s charisma and links to the symbols of royal power had held the party together, and large divisions came to the fore in his wake. After his death, the political class splintered into complex and competing factions, which cut across various lines. There were tensions between Astridians and seminarians, graduates of Astrida’s Groupe Scolaire and the Mugera seminary respectively (Chrétien and

Dupaquier, 2007:19), as well as between the ‘Casablanca’ and ‘Monrovia’ groups, each with their own orientation to international politics and the project of pan-Africanism. The former pursued a radical progressivism, while the latter had a more moderate pro-Western orientation. 196

Beyond the level of personal intrigue, there was also the continuation of the ganwa rivalry, obviously more intense following the assassination of Rwagasore (Chrétien, 2003:312).

However, all such tensions were gradually supplanted by or assimilated into the Hutu-Tutsi conflict. Following the death of Rwagasore, factions emerged around the antagonism between two figures: the Hutu Paul Mirerekano and the Tutsi André Muhirwa (Young, 2008:717). The

Hutu-Tutsi divide was introduced in Burundi by modern elites (Lemarchand, 1970:343-344;

Abrams, 1995:151; Ndikumana, 1998:34-35; Watt, 2008:30). It was also very much influenced by events in Rwanda (Chrétien, 1997:157; Lemarchand, 1994, 2009). The symbolic role of the crown was sustained following the death of Rwagasore and served to mediate the conflicts between factions for a time. The interventions of the crown led to the resignation of André

Muhirwa, and Pierre Ngendandumwe, a Hutu associated with the Monrovia contingent of

UPRONA, was made prime minister (Young, 2008:736).

Though Burundi was structurally distinct from Rwanda and the social identities therein did not have the same meanings and relationships to power, it is argued that the Rwandan revolution led to a self-fulfilling prophecy in Burundi (Lemarchand, 1994). As in Rwanda, the pro-Hutu movements received support from Belgian residents, Hutu intellectuals began to be drawn to republican ideas, and Tutsi people feared that their fate would become similar to that of Tutsi people in Rwanda (Lemarchand, 1994:60; Chrétien and Dupaquier, 2007:20). These were vital factors in the sharpening of ethnic boundaries.

Several factors intertwined to accelerate this so-called self-fulfilling prophecy. Firstly, the Rwandan revolution saw the influx of thousands of Tutsi refugees, displaced into Burundi by violence and persecution. Belgian functionaries, such as Albert Maus, saw Hutu rule in

Burundi as favourable on account of their ‘Monrovian’ anti-communist leanings, and helped to create populist ethnic parties like the Parti du Peuple (Lemarchand, 1994:60-61; Chrétien and Dupaquier, 2007:20). Another crucial factor involves the ‘neo-colonial’ overtones of the 197

Hutu revolution in Rwanda: ‘The success of the Revolution was due in no small part to the massive support it received from the Catholic church and its affiliate organizations, church- sponsored cooperatives and Christian trade unions, both closely connected with their parent organizations in Belgium’ (Lemarchand, 1994:62).

The split within UPRONA was gradually marked by violence, in particular perpetrated by the youth wing Jeunesse Nationaliste Rwagasore79 (JNR). Though they were predominantly

Tutsi, they were initially motivated more by social and political concerns than racial consciousness. In 1972, however, they were key agents in ethnic mass killing (Lemarchand,

1994:62-63). A premonition of this was seen in riots that took place in the Kamenge area of

Bujumbura in 1962. There were massacres of the overwhelmingly Hutu population here, and the Hutu membership and leadership of UPRONA were appalled. Hutu student movements based in Bujumbura emerged in the wake of this event. In 1964, the following statement appeared:

After the ignominious massacres of Kamenge, in February 1962 [by the Tutsi JRC], it would be a matter of legitimate self-defence to organise resistance, or a coup d’état, and even a revolution. We are faced with a moral obligation to defend ourselves. ‘All means are legitimate as long as one reaches the goal’, said Machiavelli, Prime Minister of Italy in the early sixteenth century [sic]. The loss of human lives is necessary and indeed inevitable. If we falter, we shall be the ones to be sent to the gallows. Do you really want to see 85 per cent of the population thrown back into slavery?… Our case is not unique in history. One needs only to read about the liberation of the Roman citizenry from the yoke of the patricians. The French, to get rid of feudalism, didn’t they have to set up gallows on 1789? Nearer to us still, in South Africa, the Boers [sic] are struggling and dying to regain their rights, and since 1959 President Kayibanda [in Rwanda] has never ceased to do likewise’ (in Lemarchand, 1970:359). Between 1963 and 1964, Burundian political conflict was centred on the antagonism between the Hutu Monrovia group and the Tutsi Casablanca group, mediated by the crown.

Tensions were heightened by external events and influences, for instance the East-West rivalries playing out on the African continent, most notably the Congo rebellion of 1963-1964

(Lemarchand, 1994:68). Tutsi extremists manoeuvred the overthrow of the Ngendandumwe

79 Later renamed Jeunesse Révolutionnaire Rwagasore (JRR). 198

government by engineering the Monrovia regime’s recognition of the People’s Republic of

China; this was contrary to the wishes of the Belgian and American governments, as well as those of the mwami (Young, 2008:736).

Eventually, a Hutu president named Joseph Bamina was installed, a candidate who unified Tutsi and Hutu elements of UPRONA. Elections were organised in 1965 and resulted in clear victories for the majoritarian Hutu parties, such as Parti du Peuple. However, the mwami named Leopold Biha, a leading Bezi courtier, as interim prime minister, which led Hutu parliamentarians to think that the crown intended to deny their electoral victory (Young,

2008:736-737). A small group of Hutu militants attempted a coup, killing rural Tutsi and burning their homes, especially in Muramvya province where the site of the royal court was situated. A brutal crackdown was enforced and thousands of Hutu peasants were killed

(Ndarishikanye, 1999:236; Young, 2008:737).

Thereafter, Tutsi hegemony was confirmed in Burundian politics, and following a period of internal turmoil the monarchy was abolished on 28th November 1966, the culmination of a campaign led by a captain called Michel Micombero (Nsanze, 2003:147-148). This also marked the end of the ganwa as a political force. As put in the declaration of the republic: ‘We proclaim that the Burundian peasants are cut from the same cloth as the princes. The Republic has made concrete the most exhilarating of utopias: to give an equal chance to everyone, to create a republican spirit means a democratic and egalitarian spirit’ (in Nsanze, 2003:153).

Another factional conflict emerged in this situation, one which had great ramifications for post-independence violence, the focus of chapter eight. Micombero was representative of a generation of Tutsi-Hima, localised in the region of Bururi, and of lesser social standing than

Tutsi-Banyuruguru who had been associated with the court in Muramvya. UPRONA became a party which essentially represented their interests. The seizing of power by this Tutsi-Hima 199

faction was a major turning point in Burundian politics (Daley, 2006:667; Young 2008:712).

I shall return to this in the next chapter.

INTERMEDIATE REFLECTIONS

To conclude, I want to make some comparative points about Rwanda and Burundi, with an emphasis on societal self-understandings. In both areas, tensions between competing societal self-understandings were expressed along lines of development that were precolonial in origin, but which had been affected in various ways by the solidifying effects of indirect rule.

In Rwanda, on the one hand was an elite group centered around the Nyginya kingdom, which saw an autonomous modernity as resting on an interpretation of precolonial sources of social solidarity prior to European encounter. Recourse was made here to the historical tradition of the abiru, and the group was explicitly anti-colonial in outlook. On the other hand, we also observed the emergence of counter-elites, who came to reject this articulation of an autonomous modernity in dialogue with the traditions of monarchical Rwanda on the basis that it was deemed to be representative of particularistic, Tutsi interests, and was emblematic of a reality inimical to the spirit of modern notions of justice and freedom, and a particularly

Catholic egalitarianism. This group was itself constituted by a tension between a movement in the south and centre of the country marked by a reformist orientation, and a restorative movement expressed by Kiga communities in the north. Both of these were expressions of anti-

Nyginya sentiment and eventually converged in a Jacobin revolutionary programme of spiritual rebirth. Any anti-European sentiment must be seen as an epiphenomenon of animosity towards the court that was seen as the most prominent of a dual colonial-imperial complex of power.

These distinctions had been racialised during the colonial entanglement.

This demonstrates some important parallels with Eisenstadt’s theorising on the Jacobin interpretation of modernity and its relationship to violence. These interpretations of the modern 200

cultural and political programmes of modernity involve attempts ‘to reconstruct the polity – to topple the old and create new political institutions on the basis of a new vision in which themes of equality, justice, freedom, and participation of the community in the political centre and process [are] promulgated’ (Eisenstadt, 1999:41). The emphasis here is on a complete break with the traditions of the past, which are associated with domination, dogma, oppression, even evil; and the increasing incorporation of protest discourses from the periphery into the core of political centre-formation (Eisenstadt, 2005, 2005a). Furthermore, these movements have a particular relationship to violence, where violence is underpinned by ‘ideological justification, amounting to near sanctification’, a justification ‘often rooted in the attempt to combine the change in symbols, bases of legitimization, and the basic institutional framework of a regime with new visions of political and social order’ (Eisenstadt, 1999:41).

The traditional Nyiginya court in Rwanda, identified with Tutsi, came to be viewed by

Hutu counter-elites as a feudal relic from which a clean break was needed (Lemarchand,

1970:133). As we shall see in the next chapter, the 1959 social revolution was the first of several episodes of Jacobin-style violence against ethnically Tutsi monarchism. These were not simply elite projects; they channelled widespread rural grievances. As Catherine Newbury

(1998:14) writes: ‘Many rural people had a stake in this new political order, and they shared a strong commitment not to return to the ‘old order’ in later years’.

Similar Jacobin-style discourses and instances of mass ‘cleansing’ and ethnic genocidal violence came to afflict Burundi after the dissolution of the monarchy in 196580. At first, however, the tensions in modern societal self-understandings crystallised around the rivalry between two princely factions, the Bezi and Batare, and the respective UPRONA and PDC political parties. The division can be located as between, on the one hand, the Batare organised

80 These are analysed in the following chapter (see pages 227-238). 201

politically in the PDC who feared a return to a ‘feudal’ precolonial Burundi and who sought collaboration with the colonial administration, and on the other, those Bezi of UPRONA who advocated an autonomous interpretation of modernity through immediate national liberation.

In the articulation of the latter, the traditions of the kingdom would be excavated and mobilised in the context of the problematiques of the transition to autonomy.

The pan-ethnic appeal of mwamiship and the crown as a symbol of nationalism, as expressed in the populism of UPRONA, mitigated Hutu-Tutsi conflict for the first three years of independence. UPRONA had a susceptibility to authoritarianism, because of the weakness of the conception of ‘the people’, and following the assassination of its charismatic leader,

Louis Rwagasore, it soon descended into infighting. The Hutu-Tutsi antagonism was a secondary conflict that was primarily carried forward by elites jostling for power within the

UPRONA party framework, thus compounding the inaccuracy of assumptions of primordial hatreds that became the prominent trope of international reportage of postcolonial violence. It was significantly influenced by events in neighbouring Rwanda and Congo, as well as the broader international context of east-west rivalries.

Finally, in the broader parameters of this thesis and in terms of its original contribution, there are three significant points raised by the analysis presented in this chapter. The first concerns the entanglements of modernity and tradition in postcolonial societal self- understandings, and their relationship to violence; the second concerns the entanglement of

Burundian and Rwandan trajectories towards independence in the broader structural problems intrinsic to global transformations of modernity; and the third concerns the entangled historical routes of Burundi and Rwanda themselves. 202

There is an important critique of the multiple modernities conception of the stability of traditional cultural programmes, which also has ramifications for contemporary debates about decoloniality and cultural revivalism. Here, it is argued that:

The key to effectively addressing contemporary problems lies in reclaiming and revitalising indigenous traditions that have been degraded and suppressed in the wake of colonialism. Colonialism violently disrupted African cultural traditions and imposed, with varying degrees of success, European forms of thought and social organization upon colonised peoples. Having achieved political independence, postcolonial Africans must now pursue a more decisive liberation, a ‘decolonisation’ of African minds and societies (Ciaffa, 2008:121). What a consideration of Burundi and Rwanda reveals is that tradition itself, rather than being dissolved or destroyed in colonial modernity, was heavily racialised, albeit upon already existing foundations. Crucially, the racialised projections of the colonisers took the traditional foundations of Burundi and Rwanda to be static and timeless and, importantly, in ruling indirectly through this notion of tradition, served to produce it in important senses. As such, societal elites, even though they might occupy a privileged position, cannot simply institute imaginaries after the model of possibilities projected into the future ex nihilo. The constraints in this instance are such that even the traditional ‘cultural programme’ to be adapted had been racially solidified under indirect colonial rule. The epistemic problematique, as discussed by

Wagner, was suffused with the coloniality of racial science; ‘societal self-understandings’ were heavily ethnicised, which as we shall see had profoundly deleterious effects in terms of mass violence (Chrétien, 2003:354-355) In sum, tradition was not something neutral that all could draw on and revive in order to utilise it in the present in the service of a better future.

In the case of Rwanda especially, tradition operated in the interests of precolonial elites.

The interests of colonisers became entangled with theirs. We have seen, in chapters five and six, how the view of a centralised traditional Rwanda, harmonious before the European colonisers because of its hierarchical structure, was itself deployed as a Nyiginya ideology (Des

Forges, 1995; C. Newbury, 1998; Newbury and Newbury, 2000). An attempt was made to 203

utilise tradition in order to nullify the egalitarian threats of an ethnicised popular republican movement that aimed to dismantle the institutional status quo and its traditional foundations.

In the framework of the Hamitic Hypothesis, the Tutsi elite were ‘destined to rule’, a sentiment that occasionally entered the realms of public discourse. The ethnicist reading of tradition came to prevail in Burundi, where it supplanted ganwa rivalries as the main vector of political conflict and undermined the monarchical sentiment that had characterised the early stages of independence movements. This ought to qualify tendencies towards what the Beninese philosopher Paulin Hountondji (1996:60) has called the ‘myth of primitive unanimity’ implicit in some decolonial-theoretical valorisations of tradition as against the neologism modernity/coloniality.

In various ways, these developments were spurred by broader institutional and cultural transformations. Cold War rivalries, the legacy of conflict in Europe, political strife in the metropole, economic and legal dependency to international institutions and powerful nations, and a new generation of Catholic missionaries, all impacted on the capacity of each country to formulate an autonomous response to political and economic problematiques. As we shall see, such forms of coloniality persisted and intensified in the postcolonial period.

Furthermore, a legacy of Burundi and Rwanda’s administrative fusing as Ruanda-

Urundi, whereupon both were restructured along the same lines despite their significant differences, can be seen in the entry of the Hutu-Tutsi conflict into Burundi. The revolution in

Rwanda led to mass displacement of Tutsi refugees, carrying traumatic memories of violence.

Moreover, the group which came to power in Burundi under Michel Micombero particularly feared an uprising of the like seen in Rwanda.

These entanglements unfolded in the post-independence period and were implicated in practices of genocide and mass violence. As Eisenstadt (2000:25-26) puts it, postcolonial mass 204

violence and genocide in Burundi and Rwanda should be conceived of not as the spontaneous and violent assertion of timeless primordial hatred, ‘but the result of the ongoing dialogue between modern reconstruction and seemingly ‘traditional’ forces’. They are to be seen as

‘traumas of modernity’.

205

CHAPTER EIGHT – POSTCOLONIAL CRISIS AND GENOCIDE, 1965-1994: TRAUMAS OF MODERNITY.

At first glance, this land just south of the equator seems a paradise far from more frenetic parts of the world. It is here where Stanley succeeded in his search for Livingston, where Bogart piloted a heap called The African Queen, where baby hippos play in Lake Tanganyika in the setting sun. But it is also here, where the green mountains rise up through clouds to the blue skies, that the tranquillity has been shattered with killings in the mist… more than 200,000 people have been butchered in tribal warfare between the haves and have-nots (Squitieri, 1993:1a)

The land is twisted and folded in on itself. Volcanoes thrust up from the horizon. Grey rivers of lava scar the plains. Earthquakes are common. Flying over this region, you quickly pass from rolling green hills to lava fields, to plains of scrub and grass, over lakes of brilliant blue and dark rivers in tropical jungles. The plane is thrown about by air currents from the jagged landscape. It is as if the heart of Africa is in turmoil. Clinging to this capricious surface are two peoples who seem to have imbibed the spirit of the land. They live and die like Siamese twins who hate each other. The Hutus and Tutsis are inextricably bound together by history, culture, language and the struggle for survival in the most densely populated countries in Africa. Hardly an inch of cultivable land is unused and the two groups live hut by hut on the same steep hills (Dowden, 1994:12) --

The above passages are fairly typical of how genocidal violence in Burundi and Rwanda during the 1990s were presented to distant bystanders in the West. In the years since the genocide, such Western media representations have been critiqued (McNulty, 1999; Carruthers,

2004, Thompson, 2007), although whether lessons were learned remains an open question81.

This is perhaps an overstated issue; discursive representations do not in themselves cause genocide, even though they have productive power (Said, 1978; Mudimbe, 1988; Hall, 1992).

It is not my intention to deal with this representational issue systematically. It is, however,

81 See the following from an Al Jazeera report on 2013 violence in the Central African Republic: ‘Parents tell of seeing their children drown or burn to death; wives tell of their husbands tied up and shot; children speak of their relatives hacked to pieces – as the failed state at the heart of Africa descends into anarchy. And from the darkness come death-squads unleashing unprecedented brutalities on civilians’ (McNeish, 2013). The cliché is hardly subtle here: one need only note the use of the words ‘heart’ and ‘darkness’ in such close proximity. It is also worth noting that Richard Dowden (quoted above), in his distinguished career as a journalist of African affairs and directorship of the Royal African Society, has done more than most to overturn Africanist clichés and highlight the diversity of the continent (2008). 206

worth briefly considering the connotations of references to topography and naturalistic metaphor deployed in these passages.

Such metaphors have a long history in the Western imagery of this part of Africa, especially during the search for the source of the Nile, which drew many European explorers into the interior of the continent. It was here that the ‘Great Lakes’ region was first evoked

(Chrétien, 2003:22, 2010). In colonial texts, nature itself often seemed to function as an explanatory device for encountering Otherness. This was a fundamentally ambivalent device, simultaneously evoking tranquillity and barbarity. What moved and horrified colonialists in

Africa, Arendt (1967:192; see also Taylor, 2004:38) wrote, was that Africans purportedly

‘treated nature as their undisputed master, that they had not created a human world, a human reality, and that therefore nature had remained, in all its majesty, the only overwhelming reality

– compared to which they appeared to be phantoms, unreal and ghostlike’. The landscape is verdant and beautiful but it is also wild and miasmic, bringing to mind Joseph Conrad’s depiction of the Congo in Heart of Darkness (1997 [1903]). The stinging criticism of Conrad administered by Chinua Achebe (1988:256) provides a lucid summation of its representative pitfalls: Central Africa is effectively written ‘as a metaphysical battlefield devoid of all recognisable humanity’.

Such devices also reveal a paradox. It is a feature of rationalised and ‘disenchanted’ modernity that we can explain and understand the processes by which volcanoes erupt and floodwaters rise. Their causes can be scientifically verified in a detached manner, without recourse to notions of divine or mythic design, and subsequently their effects on human societies can be subject to planning and intervention. However, the evocation of ‘eruptions’ of human violence and ‘floods’ of expelled peoples tend to be deployed in the absence of explanation and understanding. As Elias (2010 [1985]:87) noted, to describe human violence in terms of natural disasters is to be ‘just as helpless in the face of the terrifying dangers to 207

which people are exposed by another… as our ancestors were in the face of menacing natural forces, the thunderbolts, the plagues or the immense floods to which we owe the myth of

Noah’s Ark’.

Over the previous chapters we have developed a means of understanding these events historically and sociologically. We have seen how the tension between modernity and tradition and its institutionalisation in the colonial period both enabled and constrained the articulation of autonomous societal self-understandings on the threshold of independence. The events analysed in this chapter are inscribed onto this background. I claim that Burundian and

Rwandan genocide and genocidal violence in the post-independence period ought to be seen in the context of their specific trajectory to and through modernity. As such, I suggest that it is erroneous to locate the ‘modernity of genocide’ solely in observable set of institutional and logical components – bureaucracy, scientific racism, ‘distancing’ technologies, moral

‘numbing’ – normally following from the ideal-typical modern genocide known as the

Holocaust (Bauman, 1989).

As such, this chapter substantiates and extends S. N. Eisenstadt’s (2000:25-26; see also

2005, 2005a) intriguing but undeveloped claim that postcolonial Burundi and Rwanda comprise a ‘trauma of modernity’, the sort of which indelibly marked the twentieth century and

‘that brought into question its great promises’; that the periodic destruction and genocide which occurred in the entangled post-independence histories of Burundi and Rwanda were ‘not outbursts of old ‘traditional’ forces, but the result of the ongoing dialogue between modern reconstruction and seemingly ‘traditional’ forces’. Post-independence Burundi and Rwanda demonstrate to a tragic extent that experiences and interpretations of modernity can give rise to violent – indeed, genocidal – social conflict and trauma. 208

I will focus in particular on two events, namely the 1972 genocide in Burundi, and the

1994 genocide in Rwanda, whilst situating them within a broader historical context. Without this context, we run the risk of speaking of ‘eruptions’ of violence or evil, which are ultimately incomprehensible. The by-product, however, is a level of removal from micro-level detail, especially given the circumscriptions imposed by the form of presentation. To clarify, this does not mean that important work at this layer of social reality, for instance socio-psychological studies regarding individual motivations for participations in killing, are of lesser explanatory or interpretive value for my purposes. On the contrary, they impose important checks on the temptation to make sweeping, monocausal claims about, for instance, overpopulation, the legacy of colonialism, or the culpability of modernity tout court. Genocide is a multifarious and complex social phenomenon and my intention here is to pull out some key threads which relate to this thesis as a whole and which are of specific interest vis-à-vis the concept of

‘modern genocide’.

THE MODERNITY OF GENOCIDE

My approach here represents both a departure from and an addition to the literature on the ‘modernity of genocide’ thesis in the interdisciplinary field of genocide studies (for discussion see Scherrer, 1999; Hinton, 2002; Gellately and Kiernan, 2003, chapters 2-5;

Bloxham, 2008; Moses, 2008a). These tend – following the work of Zygmunt Bauman (1989,

1991, 1995) although, as I shall suggest shortly82, not in its full implications – to conceive of modernity in terms of ‘an era and a set of institutions’ (Wagner, 2008:8). In this sense, ‘modern genocide’ is defined in a more or less isolated and static sense, as an event occurring at an achieved state of institutional development. Modern genocide, it is said, is bureaucratic and instrumentally rational, utilises scientific racism, is biopolitical, is perpetrated by nation-states,

82 See page 212. 209

and so on, and is said to have emerged at a distinct point in time and space, developing with the rise of modernity in Europe before culminating in the Holocaust. In more recent discussions

(e.g. Stone, 2004a; Melber and Jones, 2008; Moses, 2008; Lindqvist, 2012; Kühne, 2013), the origins of modern genocide are located in European expansion in the colonies and the attendant historical developments of racism and nation-building.

The 1994 Rwandan genocide has been evoked as a rebuttal of this thesis (e.g.

Marshman, 2008; Jones, 2011:426-427). As exemplified by Benjamin Lieberman (2012:6):

It is possible to find elements of modernity in the Rwandan events: killers rifled through official records of identity to sort out Tutsis from Hutus and find victims, though they sometimes killed Hutus as well as Tutsis, and propagandists employed the modern technology of hate radio to whip up animosity. At the same time, how many of the authors who stress the modernity of killing in Rwanda would have cited the country, one of the poorest and least-developed in the world, as emblematic of modernity prior to the genocide of 1994? Does a society only become modern after it has suffered genocide? Lieberman seems to have a point. The 1994 Rwandan genocide was the most efficient among the litany of horrors in the twentieth century (Gourevitch, 1998:3). Around 800,000 people were killed in just 100 days and this efficiency was largely achieved with crude agricultural implements like the machete and small firearms (Verwimp, 2006; De Swaan, 2015:90). There were no isolated extermination camps, sites of ‘exception’ or ‘dyscivilisation’, where killing could be distanced so that ordinary life could continue unabated (Agamben, 1998; De Swaan,

2001). Killing was rarely enacted at a distance; it was often face-to-face, personalised and affective (Straus, 2008:1). To the extent that violence was compartmentalised, it occurred in stadiums, at coordinated roadblocks, in churches and schools, and in the collines on which people lived (Des Forges, 1999). In short, it occurred in spaces of everyday life83.

83 The physical geography is in part an explanatory factor in the type of killing. In Europe, there were vast, isolated spaces where death could be compartmentalised into extermination camps and put at a distance. In Rwanda, very densely populated and with the vast majority of the mountainous terrain terraced into cultivable land, this was not possible. This might be important in explaining how distant observers were so fixated on the physical and up-close nature of violence in the Rwandan genocide. 210

The first instance of genocide in post-independence Africa had occurred twenty-two years prior to the Rwandan genocide of 1994. This was the 1972 genocide in Burundi, much less memorialised than that which occurred in Rwanda (Lemarchand, 2011a:37). Here, in response to an armed Hutu rebellion in the locality of Rumonge, just south of the capital

Bujumbura, the Tutsi-Hima dominated state launched a systematic annihilation of Hutu civilians (Russell, 2015a:73). Up to 250,000 Hutu were killed, and 150,000 more were expelled from the country (Abrams, 1995:148).

In responses to questions about where cases like Burundi and Rwanda fit in relation to his writings on genocide, Bauman also affirms their differences from those he analysed. Nazi and Stalinist genocide, with their order-building imperatives, are deemed to adhere to a

‘societal’ logic of genocide. Genocidal violence in Burundi and Rwanda are said to be examples of a ‘communal logic’84 or ‘neighbourly imperialism’. Though both are ‘offshoots of the modern condition’, the latter are oriented towards community building rather than the realisation of utopian blueprints, the violent reassertion of particularistic collective identities and gemeinschaft bonds in the voids of failed states undermined by global processes (Bauman,

2008:78-119, 2010:99-107).

However, there are a number of ways in which both Burundi and Rwanda conform to the statist, ‘societal’ genocides outlined above. Genocidal violence has been made possible by an achieved level of state organisation and bureaucratic administration that extends deep into

The style of killing, the glorification of the agricultural work of the peasantry, the technology of machetes, etc. also reflects a particular position in the capitalist world system (Rodney, 1972; Wallerstein, 1973; Arnason, 1991:190) rather than outside global modernity. 84 He concedes that ‘in practice, most cases of categorical murder contain a mixture of the two, in varying proportions, and need to be plotted somewhere between the ‘ideal-typical’ extremes’ (Bauman, 2008:106) and this is indeed vindicated by Burundi and Rwanda, as I will show. Just as there were strong bureaucratic and instrumentally rational components to the genocides in Burundi and Rwanda, there were also effervescent, communal bouts of affective and emotive genocidal violence in the Nazi extermination. A number of commentators have argued that Bauman downplays the extent of this latter point (e.g. Joas, 1998:50; Stone, 2004; Moses, 2008a). 211

the lives of individuals (Reyntjens, 1987; Lemarchand, 1994; Laely, 1997). We also see a stratified and hierarchical division of labour oriented towards genocide construed instrumentally as a task, replete with concerns with efficiency, orderliness and security (Des

Forges, 1999; Fujii, 2004:99; Straus, 2008; Russell, 2015). More than this, however, they have also been collective, ‘nativist’ projects (Mamdani, 2001), involving the mobilisation of a sizable proportion of the civilian population. To borrow and extend the Bauman’s (1989, 1991)

‘gardening’ metaphor; gardeners utilise the labour of children of the soil.

Somewhat neglected in the critical appropriation of the thesis presented in Bauman’s writings on modern genocide has been the sense that genocide in the modern period entails a future-orientation. Crucially, Bauman saw that for the Holocaust to be possible, bureaucratic rationality had to be twinned with the ‘future-orientation’ of modernity, with its order-building imperative. The Nazi genocide ‘was not the work of destruction, but creation. They were eliminated so that an objectively better human world – more efficient, more moral, more beautiful – could be established’ (Bauman, 1989:92). The planners of modern genocide are akin to gardeners 85 , overseeing the implementation of a grand design and setting apart productive elements from ‘weeds’ to be destroyed.

We can see elements of this in the cases of Burundi and Rwanda. Genocidal discourse in Rwanda has expressed a modern ‘future-orientation’, a consummation of the unfinished

1959 social revolution. In Burundi, genocide has been employed as an instrument in aid of the elite mastery of crises86 (Wagner, 2015:8) and ‘reorganising’ (Feierstein, 2014) social relations so as to secure the power of a factionalised elite. In these cases, however, we have to move

85 See earlier discussion on pages 137-141. 86 Dictatorships and the disavowal of democracy, one-party states, and the social practice of genocide can, in Wagner’s terms, be seen as ‘a particular interpretation of modernity that privileges – at least within temporal limits – the mastery of political crises [postcolonial crises] over the expression of collective autonomy’ (Wagner, 2015:8).

212

beyond the level of elite planning and discourse to understand how broader populations are drawn into participation in genocide.

The narrative memorialisation of historical experiences of violence and injustice has played a significant role in the construction of collective ethnic identities, which have in turn become significant bases upon which ethno-nationalist societal self-understandings are formed.

In contexts of strife, conflict and humiliation, the work of ‘cultural trauma’ – ‘when members of a collectivity feel they have been subjected to a horrendous event that leaves indelible marks upon their group consciousness, marking their memories forever and changing their future identity in fundamental and irrevocable ways’ (Alexander, 2016:4) – can move away from appeals to ‘moral universals’ (Alexander, 2009). Instead, it turns to the ‘terror of history’ whereby ‘current events are depicted as reincarnations or perpetuations of the traumatic, often genocidal, experience’ undergone by particularistic collectivities, often leading ‘to pre-emptive or anticipatory self-defence’ (Moses, 2011:96-97).

In a related sense, Mann (2004; see also Wimmer, 2002) has noted that ‘murderous cleansing’ is a dark potentiality of democracy. In societies like Burundi and Rwanda, more-or- less bifurcated along ethnic lines (unlike the multi-ethnic neighbouring territories of Uganda,

DR Congo and Tanzania), democratisation carries the risk that the notion of ‘the people’ will be interpreted in fiercely particularistic and autochthonous ways. In certain precipitating conditions, typically conditions of crisis, the ethnos – where ‘the people’ are conceived as

‘children of the soil’, regardless of other forms of social stratification – comes to trump the demos. In Burundi, genocide has been legitimated by the perpetrators on account of a fear of the tyranny of the majority. It has historically been used by an elite class of Tutsi-Hima from the Bururi region of the country to maintain and consolidate power (Lemarchand, 1994; Daley,

2008). Genocidal violence in post-independence Rwanda was often justified by elite 213

representatives of an ethnic majority as a way of warding off a supposed return to the dark days of serfdom under ‘Tutsi feudalism’ and minority rule (Mamdani, 2001; Des Forges, 1999:144).

DISENTANGLING THE ‘MANIFOLD INTERLOCKINGS’ OF POST-INDEPENDENCE VIOLENCE

I have situated the genocidal violence of the period under investigation as traumas of modernity. In order to substantiate this claim empirically, it is now necessary to disentangle the grim histories of violence in both Burundi and Rwanda, both from each other and from the wider regions in which they are situated. Towards the end of the last chapter, we saw how

Burundi became embroiled in ethnic conflict in 1965, in no small part due to the influence of the situation in Rwanda. It is commonly acknowledged that the ‘demonstration effect’ of violence in one country in the region, in addition to the capacity of violence to send enormous numbers of people into exile with raw traumatic memories, has a causal effect on violence in another area (e.g. Uvin, 1997:105; Lemarchand, 2009; McDoom, 2010; Shaw, 2013:162-171;

Jones, 2014).

There is, however, a danger that one begins to see this purely in regional, even continental terms (see Shaw, 2013:162-171). If the trend in comparative studies of genocide has until recently been to isolate Rwanda from its regional-historical context, one should also avoid the temptation to see it purely in terms of this regional-historical context. As David

Newbury (1998:75; see also Lemarchand, 2009:4) writes of the Great Lakes region and its turbulent postcolonial history:

It is tempting to see this recent history simply as ongoing regional turmoil, with tragedy spilling from one country to another. However, though mutually reinforcing, these conflicts were not simply the extension one of the other. Instead, each had its own history. Despite the fact that political tension and violence in each country clearly exacerbated violence in the others, each one also had deep local roots; they were ‘convergent catastrophes’, independent in origin, even as they were interdependent in their evolution. It is therefore useful to retain sight of the historical roots of such crises, to disentangle their manifold interlockings, if we are to understand them. 214

The following is an attempt to ‘disentangle the manifold interlockings’ of post-independence violence in Rwanda and Burundi. As I document this history, I will bring in and elaborate on insights that I provided in the previous section.

RWANDA

THE FIRST REPUBLIC In the case of Rwanda, it is necessary to begin in the aftermath of the revolution. In the ensuing years there grew a sense of disappointment at the Kayibanda government. Internal rivalries intensified between the northern and central constituencies of PARMEHUTU, and there emerged a great tension between the regime’s emphasis on revolutionary aims and the lack of resources available with which to implement them (Lemarchand, 1970:240; Reyntjens,

1987:91). As a 1968 regime report stated (in Reyntjens, 1987:92) very critically:

Unity, concord, mutual support, confidence, co-operation, patriotism all have lost their meaning and have ceased to exist. These values have been replaced by condescension, hatred, egoism, antagonism, dishonesty, race for money, dissension and regionalism. The popular masses feel that their leaders have betrayed them when saying that the revolution of 1959 would free them from injustice. They realise now that this was merely a way of getting hold of functions; once these were occupied, injustice became worse than before. As in Burundi and elsewhere in Africa, the state was patrimonial and prone to corruption and nepotism, used to enrich elites (Uvin, 1998:21; Bayart, 2009). Yet Kayibanda’s regime legitimated itself by making recourse to the social revolution and its legacy. In this schema, the Hutu had wrenched power away from the ‘feudal’ Tutsi of the Nyiginya court and established a true republican and majoritarian democracy (Lemarchand, 1970:255; Reyntjens,

1996:246; Uvin, 1997:97-98).

Tutsi people were a convenient scapegoat group onto which the disappointments of the first republic were cast. Ever-present at this time was the prospect of an invasion of Tutsi refugees into Rwanda, who included among them factionalised UNAR elite and the exiled mwami (D. Newbury, 2005:270; Young, 2008:734; Long, 2012:213). They called themselves 215

inyenzi87. These insurgents aimed to overturn the transformations of 1959-62. At the end of

1963, an attack was launched from Burundi by some two thousand Tutsi fighters. After a short, intense battle, they were defeated, many having been killed and others having fled into the bush.

PARMEHUTU used this to reinvigorate the party and to create a sense of ethnic solidarity amongst the Hutu majority population. The leaders of UNAR who had remained in Rwanda, among other Tutsi political elites, were executed, despite having broken their allegiance to the monarchy, declared their commitment to the republic, and strongly condemned the refugee incursion.

The counter-insurgent Tutsi were conflated with a supposed Tutsi ‘fifth column’ inside

Rwanda and in 1963 the latter were subject to massacres. These were organised killings, overseen by burgomasters and prefects, supervised by ministers appointed by the government, and legitimated by recourse to ‘self-defence’ (Chrétien and Dupaquier, 2007:19). Civilians were rallied to the cause, manning roadblocks and partaking in massacres. This marked an important difference from the Jacobin violence of the revolution: ‘1959 saw few outright killings as part of a political conflict; 1963 saw killings on a massive scale in response to military intervention’ (D. Newbury, 2012:51). The international community took little interest in what was deemed a matter internal to Rwanda, though some high-profile observers like

87 Inyenzi translates as ‘cockroach’, which gives important context to the widespread use of the term ‘cockroach’ in the genocidal discourse of 1994 (Des Forges, 1999:62). At first glance, it is tempting to view such vocabulary – very common in the hate radio of RTLM and in the pages of Kangura – as a fairly straightforward term used to describe the sub-human, pestilent and unwanted character of a victim-group conceived as a threatening presence to the body politic. Such language is common in modern genocide (Savage, 2007). However, in Rwanda it has a meaning traceable to the formation of this group exiled after the 1959-62 social revolution. It was a self-appointed name for an insurgency that ‘travelled at night’ and ‘refused to go away’, something that the hardiness and survivability of the cockroach symbolised (Nyakabwa, 2002:84). At numerous points in the post-independence history of Rwanda, notably during 1963 and 1966, this group attempted to invade Rwanda. Given that such an invasion had begun in 1990 (by which time the group were called inkotanyi after the warriors of Rwabugiri’s Nyiginya kingdom (Des Forges, 1999:61), and inyenzi had retained only its pejorative connotation) and was still alive in 1994, the use of ‘cockroaches’ has a significance in the Rwandan context not contained within the notion that it describes only the subhumanity of the victim, even if that is certainly part of its meaning. It is a term deeply associated with monarchism and republicanism. 216

Bertrand Russell (1964) drew attention to the systematic murder by likening it to the Holocaust.

The Rwandan government communicated to this ‘outside world’ that the massacres were the result of tribal passions run amok, which they were unable to stop. These events, said to have been responsible for 15,000 deaths and over 100,000 displaced people, constituted an ominous foreshadowing of the 1994 genocide. (Segal, 1964:13-20; Baker, 1970; Kuper, 1970:372-374;

Chrétien, 2003:306; D. Newbury, 2005:272).

THE SECOND REPUBLIC In spite of their momentary effervescent homages to the revolution, events such as those outlined above nevertheless contributed to economic strife, and Rwanda became increasingly isolated in the proceeding years (Baker, 1970:148; De Lame, 2004:293). The cleansing of Tutsi from civic positions, educational institutions and so on, that had begun during the revolution, continued (Uvin, 1997:101-102). In this period, violence in Rwanda was often influenced by events in neighbouring countries. In 1972, during the genocide in Burundi, PARMEHUTU published a document stating that ‘Tutsi domination is the origin of all the evil the Hutu have suffered since the beginning of time. It is comparable to a termite mound teeming with every cruelty known to man’ (in Chrétien, 2003:307). This general decay prompted a coup in 1973, led by Juvenal Habyarimana, and the establishment of a Second Republic, which lasted until

1994 (Reyntjens, 1986; Jefremovas, 2002:3). The transition from the First Republic of Grégoire

Kayibanda to the Second Republic of Juvenal Habyarimana is perhaps most notable for the fact that it transferred power from south-central elites to the northern Hutu people, who retained a degree of cultural autonomy from southern and central Hutu groups owing to their late absorption into the state of Rwanda. This transition of power was aided by colonial powers

(Pottier, 2002:15; De Lame, 2004:293).

The key word in the discourse of the Second Republic was ‘development’, an image which it projected successfully both externally and domestically (Uvin, 1997:97). 217

PARMEHUTU was abolished, the single political party in control of the state was renamed the

Mouvement Révolutionnaire pour le Développement (MRND) (Uvin, 1998:24; Chretien,

2003:308). Development was to be ‘auto-development’ and Rwanda was to become self- sufficient, ‘to live within its means’ (Verwimp, 2000:336, 2013; Uvin, 1997:99). The notion of development, a modern societal-self understanding, was founded on the hard graft of the peasantry, the sons and daughters of the soil. The coup d’état and the birth of the Second

Republic was, in Habyarimana’s words:

Above all a moral coup d’état. And what we want, and we would consider our action as failed if we do not reach this goal, what we want, is to ban once and for all, the spirit of intrigue and feudal mentality. What we want is to give back labor and individual yield its real value. Because, we say it again, the one who refuses to work is harmful to society (in Verwimp, 2000:335). In spite of its Arcadian and romantic veneer, this was very much a future-oriented societal self-understanding with patent biopolitical connotations. It was also highly exclusivist.

Influenced by colonial-era stereotypes about Bantu agriculturalists and the Hamitic ruling class,

Tutsi did not figure in this schema due to the perception, as Mann (2004:445-446) notes, that

‘work was seen as a defining quality of the ‘Hutu nation’. Hutu people were ‘tough’, ‘solid’, in contrast to the parasitic Tutsi people, whose ‘delicate constitutions’ were unsuited for labour.

‘They cannot do painful chores’ or, in more proletarian ideology, ‘They eat our sweat’’.

Rwanda became something of a ‘darling’ of the international development community in this period (Anderson, 2000; Jefremovas, 2002:1). In terms of the outside world projection, in a survey of World Bank reports, Uvin (1998:40-50) notes that they frequently mention

Rwanda’s political stability, its effective administration and organisation, and its sound management, as well as its concern for the rural population. One report even noted ‘the cultural and social cohesion of its people’, failing or refusing to acknowledge the continued stigmatisation and marginalisation of the Tutsi population (Uvin, 1998:44; Nyakamba,

2002:157-158; De Lame 2004:283). 218

During this period, the trend towards centralisation accelerated rapidly. With the support of development enterprises, as Uvin (1997:97) notes, the state was:

able to expand its presence to the most remote corners of the territory and of social life. Representatives of the state and of the single party were present at even the lowest level of social organization: each ‘colline’… each extended family was in permanence surrounded by centrally-appointed administrators, teachers, agricultural monitors, internal security agents or police agents, as well as by local party cadres of all kinds. The state was in charge of all fields of human endeavour, from education, health and rural development, to the promotion of culture and the ‘right’ social values – much of this financed by vast quantities of development aid. This period was also characterised by a more tolerant interethnic relationship than was experienced during Kayibanda’s First Republic. While Hutu people dominated political positions and state and development funds, Tutsi people still retained an influence in the non- state private economy and in the development sector (Uvin, 1998:211; Mann, 2004:438; D.

Newbury, 2005:272). Nevertheless, lines of division remained, including and in addition to the

Hutu-Tutsi split. Southern Hutu groups were excluded from power and there were instances of unrest in the south at the end of the 1980s, when the economic crisis was in full swing (Uvin,

1997:94).

The development projected externally and utilised internally was stalled by a series of external shocks. The collapse of world commodity prices after 1985 led to a fifty percent drop in the price of coffee by the turn of the 1990s, and other primary exports in the country, like tea and tin, also fell in this period (Kamola, 2007). Coffee accounted for 75-80 percent of export earnings in 1990 (Economic Intelligence Unit, 1993:61). Combined with the steep rise in poverty were a number of ecological factors which further impacted economic activity, such as the impact of droughts in the 1980s and the effects of land degradation, erosion and dense human settlement (Uvin, 1997:106). Rwanda was forced to accept a structural adjustment programme from the International Monetary Fund and the World Bank in order to qualify for loans: ‘From 1985 onwards, millions of Rwandans saw their misery increase dramatically, while their prospects for the future disintegrated’ (Uvin, 1997:107-108; see also Storey, 2001; 219

De Lame, 2004; Mann, 2004:439-440). As conditions in the country deteriorated, alliances with development brokers became crucial. By the 1990s, Rwanda was the largest recipient of aid from both Belgium and Switzerland, and received substantial development funds from the

UK and USA (Uvin, 1998:42).

Such factors contributed to significantly undermining the glorification of the peasantry that was central to MRND legitimation. The disappointments and frustrations of the population became humiliation. Dissatisfaction with this, and its anomic effects, was expressed by

Emmanuel Ntezimana (in Longman, 2011:125), one-time professor at the National University of Rwanda:

Very curiously and quite dangerously, in a very short lapse of time, mutations and ruptures have been produced in Rwanda, notably at the level of ‘ubupfura’ (dignity), ‘ubugabo’ (courage) and ‘ubudahemuka’ (integrity). There have been sudden transformations operating in the name of ‘development’. Among the elite and people of note, the goal of life, the value of the individual and the quality of society seem henceforth reduced to the search and fantastic accumulation of goods and material comforts, at the same time delicious and poisonous. And all means appear good, including those falsely called ‘political’. This poorly thought out seeming ideology of ‘development’ disarticulates and destroys the substratum of our society. It is in its name that, in the absence of a secondary sector, an elitist minority can monopolize the machinery of the state to manipulate and marginalize, otherwise said to exploit and colonize, the popular masses.

CIVIL WAR Into this environment came the invasion of the Rwandan Patriotic Army (RPA), the armed wing of the Rwandan Patriotic Front (RPF). The group was comprised of Tutsi exiles in

Uganda, who had been displaced by expulsions at various points in post-independence history, beginning with the 1959-61 violence associated with the revolution (Reed, 1996; Kuperman,

2004). They had been caught up in the intense factional politics of the Idi Amin and Milton

Obote regimes in the 1970s and 1980s (Mamdani, 2001:159-184; D. Newbury, 2005:273), and had fought to aid President Yoweri Museveni’s ascent to power in 1986 (Nyakabwa, 2002:99).

Even the most integrated among these refugees in Uganda were ‘overwhelmed by feelings of 220

rootlessness and loss88’ (Watson, 1991:9), and there proliferated among them a narrative in which future-orientations entailed ‘returning home’ (Newbury, 2005) to the homeland from which they had been driven following the social revolution.

On 1st October 1990, the RPF launched an insurgency from the south of Uganda which materialised into a civil war (Amnesty International, 1992). The invasion was used by the

Habyarimana government as a pretext for entrenching power, explicitly instrumentalised by the ruling elite. They even staged military events to ramp up fear and to justify repression against Tutsi notables. As the following details:

On the night of October 5, 1990, a simulation of an attack on the city of Kigali and consequent combat between the aggressors and the Rwandan Armed Forces was organised by the regime. Gunshots of all types were heard throughout the night in the various districts of the city. Today, it is commonly admitted that this mock invasion was organised to allow those in power to round up certain individuals, both in Kigali and throughout the country, either on the basis generally of their ethnicity, or their manifest opposition to the government. The pretext for rounding up such individuals was that they were collaborators or potential accomplices of the enemy (Twagiramutara, 1998:117). A state of emergency was broadcast in the morning over the national radio and was in place for four days, during which time many people thought to be accomplices were rounded up, imprisoned, tortured and killed (Amnesty International, 1992).

This period also saw the rise of an infamous hate media, principally in the form of the radio station RTLM and the printed publication Kangura (Schabas, 2000a; Frère, 2007:79-111;

Thompson, 2007). In both media, in addition to commentary on the situation in Rwanda, one can find many analyses of the situation in Burundi, demonstrating the extent of the influence that events in one had on the other89. They were also significant tools in the production and

88 Watson (1991:9) quotes from a number of refugees, including the following: ‘You change your name, you become meek, you lose yourself as a person, you hide away from your culture’; ‘I’d be free if I went there [to Rwanda]… I’d build a big house. Here, we are living in a small one… There is good life in our country’; ‘I have never been with Hutu… I just hear the stories about what they did to us. How they killed our grandparents and elder brothers with knives’. 89 As demonstrated in the radio transcripts and the scanned newspaper images held in the archives of the Montreal Institute for Genocide and Human Rights Studies, available online at http://www.concordia.ca/research/migs/resources.html. 221

dissemination of a Hutu republican ideology. For instance, in its December 1990 issue,

Kangura published a widely read article which laid out the Hutu ‘Ten Commandments’, which were as follows (in African Rights, 1995:42-43):

1. Every Hutu should know that a Tutsi woman, whoever she is, works for the interest of her Tutsi ethnic group. As a result, we shall consider a traitor any Hutu who

 marries a Tutsi woman  befriends a Tutsi woman  employs a Tutsi woman as a secretary or a concubine. 2. Every Hutu should know that our Hutu daughters are more suitable and conscientious in their role as woman, wife and mother of the family. Are they not beautiful, good secretaries and more honest? 3. Hutu women, be vigilant and try to bring your husbands, brothers and sons back to reason. 4. Every Hutu should know that every Tutsi is dishonest in business. His only aim is the supremacy of his ethnic group. As a result, any Hutu who does the following is a traitor:

 makes a partnership with Tutsi in business  invests his money or the government's money in a Tutsi enterprise  lends or borrows money from a Tutsi  gives favours to Tutsi in business (obtaining import licenses, bank loans, construction sites, public markets, etc.). 5. All strategic positions, political, administrative, economic, military and security should be entrusted only to Hutu. 6. The education sector (school pupils, students, teachers) must be majority Hutu. 7. The Rwandan Armed Forces should be exclusively Hutu. The experience of the October 1990 war has taught us a lesson. No member of the military shall marry a Tutsi. 8. The Hutu should stop having mercy on the Tutsi. 9. The Hutu, wherever they are, must have unity and solidarity and be concerned with the fate of their Hutu brothers.

 The Hutu inside and outside Rwanda must constantly look for friends and allies for the Hutu cause, starting with their Hutu brothers.  They must constantly counteract Tutsi propaganda.  The Hutu must be firm and vigilant against their common Tutsi enemy. 10. The Social Revolution of 1959, the Referendum of 1961, and the Hutu Ideology, must be taught to every Hutu at every level. Every Hutu must spread this ideology widely. Any Hutu who persecutes his brother Hutu for having read, spread, and taught this ideology is a traitor.

This final aspect, the fear of the reversal of the 1959 revolution, was pervasive in the

RTLM broadcasts, in the pages of Kangura and in government rhetoric. Indeed, republicanism and its intertwinement with ethnonationalism is an underdeveloped theme in the literature on

Rwandan postcolonial history and genocidal violence. Demonstrations were organised in 222

November 1990 to protest against any attempt to reinstitute the old ‘feudal’ regime. Protestors carried placards bearing such slogans as ‘‘Let slavery, servitude and discord be finished forever!’ ‘We condemn the exploitation and servitude of the people!’ ‘Long live the republic!

Down with the monarchy!’ ‘No more feudalism! No more Kalinga!’ [the drum that symbolized the power of the ruler]’ (Des Forges, 1999:64).

Such sloganeering extended beyond political discourse into popular culture. It is pervasive, for instance, in the lyrics of the popular singer Simon Bikindi, whose songs were played regularly on RTLM. To take his song Twasezereye Ingoma ya Cyami (We Said

Goodbye to the Monarchy) as one example:

Remember the whip and the chore! Remember the days you spent Serving the master without remuneration And therefore rejoice of Independence!

We said goodbye to the monarchy The feudal and colonial yokes disappeared at the same time And we got the democracy that suits us. Come and let us celebrate the independence (ICTR, 2008; see also Des Forges, 1999:64).

It was during this period that a group called the akazu, meaning ‘little house’, came to prominence. This was a collection of people close to president Habyarimana, headed principally by his wife Agathe Habyarimana, who were representative of elite clan lineages in the north of the country (C. Newbury, 1998:15; Des Forges, 1999:40). They are largely seen as the architects of genocide, the plans for which have been argued by some to have materialised in the civil war period (e.g. Des Forges, 1999; Melvern, 2006). MRND, under the influence of the akazu and threatened by the power-sharing agreement, created youth groups and trained them in military combat, most infamously the interahamwe – ‘those who stand together’ – and another extremist Hutu party, the CND, formed the impuzamugambi – ‘those with the same goal’. The memberships of these militias were drawn largely from the swathes 223

of unemployed and immiserated youth in the urban centres of the country, whose opportunities for advancement had been crushed by the multiple crises of the late 1980s and early 1990s.

There were several Hutu refugees from Burundi among their ranks (Des Forges, 1999:283;

Mann, 2004:462; Fletcher, 2007)

At the same time, negotiations were held between the Rwandan government and the

RPF, significantly encouraged by international observers, and an interim power-sharing agreement known as the Arusha Accords was signed in 1993 (Mann, 2004:439; Newbury,

2005:276). Nevertheless, the tension ramped up and society was incrementally militarised.

(Mann, 2004:445-446). In 1993 large quantities of firearms, machetes and other agricultural implements were imported using foreign aid funds, notably from the governments of Egypt,

South Africa and France, and distributed to the interahamwe and impuzamugambi (Human

Rights Watch, 1994; Mann, 2004:446). There were several ‘practice massacres’ in various precincts in 1992 and 1993 (Human Rights Watch, 1993).

GENOCIDE On 7th April 1994, rocket fire brought down a plane containing President Habyarimana, along with a contingent which included the Burundian president, Cyprien Ntaryamira. The responsibility for the assassination of Habyarimana remains contested; there are versions of the event that implicate the RPF and others which point to Hutu extremists (Des Forges, 1999:145;

Lemarchand, 2009:xii). In the uncertainty that followed, and with strong allusions to security and maintaining law and order, the National Guard coordinated by the akazu set about implementing genocide. Among the first to be killed were political opponents and moderates; alternative voices and options were shut down in the context of crisis (C. Newbury, 1998:16;

Des Forges, 1999:144-175). 224

Rwanda has an historical legacy of forms of communal labour, extending to the still- practiced tradition of umuganda, and through the colonial corvée system of uburetwa. In the post-independence period, these traditions came to be utilised as tools of state-building

(Uwimbabazi and Lawrence, 2013). As burgomasters had previously used their authority to summon citizens for communal projects to end illiteracy and vaccinate children against diseases, they now coordinated participation in such activities in massacring Tutsi (Des Forges,

1999:182; Fujii, 2004:108). RTLM (1994, April 11) announcers told listeners that ‘this war cannot be fought by the army only. The population have to protect themselves’. As Des Forges

(1999:191) states:

The announcers replayed all the now familiar messages of hate: the inherent differences between Hutu and Tutsi, the numerical superiority of the Hutu – the rubanda nyamwinshi [the great masses], the majority people – the cleverness of the Tutsi in infiltration, their cruelty, their cohesiveness, their intention to restore past repression, the risk they posed to the gains of the 1959 revolution, and, above all, their plan to exterminate the Hutu. In a speech made a week into the genocide, the newly-named interim president,

Théodore Sindikubwabo, spoke of the need to maintain ‘peace in the hearts of our citizens so that they will be tolerant of each other and pardon each other’ and urged Rwandans to ‘keep calm, to forget all feelings of anger, hatred or vengeance’, while also urging them to ‘denounce any person who still has the evil intention of making us return to the situation of the past’ (in

Des Forges, 1999:196). Killing was referred to as akazi – meaning ‘work’ – in both official pronouncements and in the testimony of perpetrators, a word with special connotations that extended back to the ‘work of violence’ in the revolution of 1959-61 (Prunier, 1995:138-142;

Des Forges, 1999:194-195, 364; Hintjens, 2001:39; Hatzfield, 2003:60-65).

The Rwandan genocide is especially notable for the extent of civilian participation.

Scott Straus (2004), in the most reliable estimation, put the number at around 200,000.

Assisting the Presidential Guard and the death squads of the interahamwe and impuzamugambi 225

were people drawn from collines, whose reasons for participation were multifarious90 (Des

Forges, 1999: Hatzfield, 2003; Mann, 2004: Straus, 2008). They were drawn from all sectors of Rwandan society, though individual labour was stratified according to class and status.

Where poor Hutu workers were likely to man roadblocks and engage in hunts of escapees in marshlands (see Hatzfield, 2003), members of the clergy, business owners, university professors and doctors arranged massacres (Longman, 2011). It is in this context that relations between ‘ordinary people’ – neighbours, friends, professional acquaintances – facilitated large- scale violence. As put by a survivor of the genocide, Rosette Sebasoni91:

The worst thing I saw was how our parents were killed by people who knew them, by their own neighbours; people who used to come and eat home, people who were once taken to hospital by our parents, people who once lived at home were the same people who came in the attack to kill them. And killed them with a painful death, that is the thing that hurts me the most. Death is a common thing, but being killed by people he never thought would ever kill him! He was good to them and everyone used to say so, but when things changed they all came to kill him.

The Rwandan genocide has provoked much discussion and has provided details for an array of grand historical treatises: on the ‘coming anarchy’ threatened by global overpopulation, deregulation, scarcity, crime, tribalism and disease (Kaplan, 2002); on violence, with themes ranging from its permanence in the biological constitution of human beings to the frail historical civilising project of its decline (Wrangham and Peterson, 1996; Pinker, 2011); on the evils of organised religion in popular atheist tracts (Hitchens, 2008:173-195); and on the imperialist hypocrisy of US foreign policy interventions (Herman and Peterson, 2010:51-68).

In spite of their ambition, such brief allusions to the 1994 genocide can often be overly reductive or essentialist, perpetuating the popular and monocausal explanations for the

Rwandan genocide that focus on its ‘failed statehood’, the attachment of Rwandans to

‘traditional’ values of conformity and obedience, the ‘ancient tribal hatreds’ between Tutsi and

90 Luke Fletcher (2007) notes how participants outside the militias often referred to the conversion in terms of ‘turning interahamwe’ (see also Hatzfield, 2003). 91 From Genocide Archive Rwanda http://www.genocidearchiverwanda.org.rw/index.php?title=Kmc00008/kmc0008vid_1 226

Hutu, or a deadly combination of poverty and overpopulation (e.g. Gros, 1996; Diamond,

2010:311-328).

The most reliable and pithy commentary, that which grapples with the event in its complexity, of course comes from those whose scholarly careers have been dedicated to understanding Rwanda and the wider Great Lakes region. For this reason, I will conclude my commentary by referring at length to two of them:

Mobilizing thousands of Rwandans to slaughter tens of thousands of others required effective organization. Far from the ‘Failed State’ syndrome that appears to plague some parts of Africa, Rwanda was too successful as a state. Extremists used its administrative apparatus, its military and its party organization to carry out a ‘cottage-industry’ genocide that reached out to all levels of the population and produced between five hundred thousand and one million victims (Des Forges, 1995:44)

This was genocide, not an anarchic effect of ‘popular anger’ provoked by the death of the ‘father of the nation’ or by an ‘interethnic quarrel’ – arguments that the interim government used with some success. These massacres stemmed from a deliberate choice by a modern elite, and they convey the Rwandan state’s capacity to manage and mobilize the population. The organizers included high-ranking military officers; leaders of political groups that subscribed to the logic of Hutu Power; senior administrative officials, who organized not only the gathering of victims in stadiums, churches, and schools, which turned into slaughterhouse after slaughterhouse, but also the redistribution of the victims’ goods and the preparation of mass graves; and gendarmes, who flushed the human game by threatening it with guns, thereby facilitating the work of militias equipped with machetes. This decentralized machinery also included diplomats charged with justifying the killings inside the UN; bishops who affirmed their support for the ‘interim government’ or refused to answer appeals for aid from terrorized schoolchildren; religious officials who called for ‘security’ meetings, organized regroupments but gradually skimmed off some victims for the militias, and slowed down the evacuation of those in danger; academics who drafted disinformation; doctors who rushed to purify their hospital; teachers who worried about ‘order’ in their schools; and journalists who denounced the ‘cockroaches’ and rejoiced at their death on RTLM’s airwaves… These well-dressed assassins let the peasants dirty their hands, but they were behind the front lines coordinating events. In this atmosphere, several thousand militiamen initially did the work, but they little by little recruited the masses to man the patrols and the barriers and to join in mobs, in which people covered themselves with banana leaves, around slaughter sites. The killers participated in communal tasks, in ‘work’ (akazi) that, if not finished at night, was completed the next morning. They exterminated ‘cockroaches’ and ‘rats’, and they ‘swept dry banana leaves before burning them’ (Chrétien, 2003:332). We will return to the Rwandan genocide in the concluding section of this chapter, whereupon the analysis will be oriented towards the theoretical themes raised in the beginning of this chapter. Comparisons will also be made with events in Burundian history, to which I now turn. 227

BURUNDI

FIRST REPUBLIC We saw towards the end of the last chapter how the Hutu-Tutsi split came increasingly into Burundian politics following independence, initiated by modern elites, and how the court mediated between these conflicting groups. The much more longstanding competition between ganwa factions also played out in this environment, but decreased in salience shortly afterwards.

As we have seen, the incipient ethnic conflict came to a head in 1965, with a failed coup and a brutal repression (Ndikumana, 1998:35-36; Mariro, 2005). In 1966, the court was overthrown and Burundi became a republic governed singularly by UPRONA. This military seizure of power was led by Captain Michel Micombero.

The new regime couched its victory in the Jacobin language of anti-monarchism and revolution. As Lemarchand (1994:78; see also Chrétien and Dupaquier, 2007:25) details:

‘Unity and Revolution’, the magic words of the First Republic, set the tone for the daunting tasks that lay ahead. Not only the restructuring of the state but the regeneration of man himself emerged as the central objective of the Micomberist revolution. ‘We want to create and give to the world a new type of revolutionary whose character traits will fit the motto of our party, a man for whom the notions of Unity, Work and Progress are fundamental’. Only through a drastic redefinition of state-society relations could this new form of humanity come into existence. From the ashes of the monarchy, a new state apparatus would take shape, one forged on the anvil of the revolution and one through which unity would be restored and the wheels of progress set in motion. New institutions were established, like the Conseil national révolutionnaire (CNR), which was designed to protect the ‘revolution’, and ‘integral movements’ were formed such as the Union des travailleurs du Burundi (UTB), the Union des Femmes burundaises (UFB) and the Jeunesse révolutionnaire Rwagasore (JRR). UPRONA’s constitution established itself as the principal party of the republic, presiding over a single party state and the institutions of civil society, economic activity, student movements, the military and the Church (Chrétien and

Dupaquier, 2007:25-26). 228

A new elite emerged, explicitly opposed to the monarchy and united by a ‘strong reformist zeal and a concern for secularism, social progress and technical innovation’

(Lemarchand, 1970:404). They were comprised principally of intellectuals, youth leaders, important bureaucrats and military figures. The tone of their rhetoric was stridently anti- colonial and was suffused with pan-Africanism. The Union Nationale des Etudiants du Burundi (UNEBA), for instance, combined their fierce criticism of the monarchy92 with support for ‘all the movements and all the organisations struggling for the development, dignity and welfare of mankind; of all the peoples who struggle against the yoke of foreign domination; all of the movements which combat racial, social, religious, and political discrimination; of all that is likely to hasten the emergence of a United States of Africa’ (in Lemarchand, 1970:406).

In Burundi itself, UNEBA were devoted to the reconstruction of society and to the dismantling of the forces of tradition, publishing extensively in journals like Remarques

Africaines. As demonstrated in a widely-read JNR document (in Lemarchand, 1970:407-408) entitled La Jeunesse après l’Indépendence:

In our minds, this independence which we have today acquired has never meant anything else than a fundamental ‘rite de passage’, a transitory liberation and not an absolute denouement through which everything is settled once and for all. The essence of independence is what is at stake in our present and future combat. Our combat will go on forever, because the real meaning of independence will always be called into question. Our country is in a state of full evolution… We are witness to the decolonisation of an entire nation, and we want this decolonisation to penetrate the innermost recesses of our souls, not merely the surface. Independence must be a creative destruction… We have faith in ourselves and we place great hopes in our youthful vitality. We want to educate ourselves and we know that, next to our daily bread, education is our first concern. We believe that we shall succeed, because the restraining factors of colonialism have now disappeared. Independence must mean first of all the destruction of the forces which hold back the development of the country, its economic progress, its social, human and cultural development. But the problems with which we are faced are not only economic in nature… we must also create new ways of doing things, a new way of life, a new way of

92 Anti-monarchist sentiments among students and intellectuals were inflamed by their experiences during the period traced in the previous chapter, when the court in the immediate post-independence movement played a ‘mediating’ role in an increasingly factionalised political scene. Its ‘meddling’ in the political affairs of the country was interpreted as contrary to the modern commitment to autonomy. As a notable figure in the student movement suggested to René Lemarchand: ‘Isn’t the raison d’être of an intellectual to create? And how in the world can we create when the court insists on making decisions for us?’ (in Lemarchand, 1970:406-407).

229

thinking. In order to build a nation, and at the same time safeguard the liberty of the people and individual freedom, each one must actively participate in the building of the nation. Youth knows that anything can be changed, and it knows it can do so. To meet the needs of our country will be the responsibility of our youth.

However, these discourses were out of step with the capacity of the First Republic to mobilise the population, as they had been during the transition to independence around the symbols of the now-abolished monarchy (Chrétien and Dupaquier, 2007:23-25), and with its ability to heal the wounds of the 1965 repression. The economic situation of the country was also dire. Reminiscent of the crisis of late 1980s in both Burundi and Rwanda, a slump in the global price of coffee – which accounted for 80% of the country’s export revenue – in the years between 1967-69 contributed significantly to strife among the rural peasantry, who felt the subsequent recovery of the early 1970s particularly sharply and grew resentful of the nepotism of UPRONA and its failure to turn tax revenue into public services and educational opportunities in the countryside (Chrétien and Dupaquier, 2007:31-34).

Furthermore, UPRONA itself was extremely fragmented from the very establishment of the republic, in spite of the rhetorical emphasis on national unity (Chrétien and Dupaquier,

2007:22-23). Between 1969 and 1972, Micombero came more and more to represent the interests of what became known as the ‘Bururi lobby’, named after the southwestern province of his origin. This faction, who progressively controlled the military and political and administrative institutions, tended to be young (Micombero was just 26 when he assumed power) and from downtrodden social backgrounds (Lemarchand, 1994:80; Ngaruko and

Nkurunziza, 2005:41-42; Chrétien and Dupaquier, 2007:22, 51-54). They were defined in the somewhat essentialist terms of missionaries and other outside observers as Tutsi-Hima, or

‘Tutsi of the south’ (Chrétien and Dupaquier, 2007:52-53). Indeed, the Tutsi-Hima in Bururi might be compared with Bakiga region and northern Hutu groups in Rwanda, who assumed power with the advent of Juvenal Habyarimana’s Second Republic. Tutsi-Banyaruguru, the

‘northern Tutsi’, as they were known by colonial writers, conversely came from clan lineages 230

with close links to the monarchy. They were particular associated with the province of

Muramvya, the site of the former royal court where royalist sentiments were strongest among

Tutsi and Hutu residents alike (Lemarchand, 1994:81; Prunier, 1995a:9).

The intra-ethnic factionalism paradoxically had a precipitating effect on Hutu-Tutsi antagonisms. The amplification of a Hutu threat tended to be a source of Tutsi solidarity across factional lines. The fear was that Hutu people would rise up in a Rwanda-style revolution, in this case reinstalling the monarchy rather than abolishing it, and annihilate their Tutsi compatriots through political and/or physical means (Lemarchand, 1994:100-101; Chrétien and

Dupaquier, 2007:21, 108). Tutsi people with links to the monarchy were also viewed with mistrust; the higher the social ranking (based on clan lineages, ganwa links, region), the greater the chance of being linked to a monarchical plot in the eyes of the Bururi group (Lemarchand,

1994:87). The military was purged of Hutu soldiers in a particularly tense period in 1969, and there were also attempts to purge the army of Banyaruguru (Nsanze, 2003:210; Chrétien and

Dupaquier, 2007:3939-50). The exiled former mwami Ntare V returned in obscure circumstances, was accused of plotting to reclaim rule and dissolve the republic, and was subsequently arrested (Chrétien, 2003:315). Such events were crucial in fermenting an atmosphere of fear and crisis. Hutu elites came to see armed rebellion as the only viable response (Lemarchand, 1994:87).

GENOCIDE 1972 On 29th April 1972, a Hutu uprising began in the southern provinces of Rumonge and

Nyanza-Lac, assisted by Mulelist93 rebels from eastern DR Congo (Chrétien and Dupaquier,

2007:97-107). Around a thousand Tutsi were killed in the countryside (Russell, 2015a:76). In

93 Mulelists were fighters under the Congolese rebel Pierre Mulele, deceased before 1972. Mulele was the leader of the Simba rebellion in DR Congo (then Zaïre) in 1964. About 25,000 Zaïrian refugees were present in southern Burundi at this time and, largely as a result of discrimination, shared the animosity to the Tutsi-Hima regime (Lemarchand, 1980:55; Russell, 2015a:80). 231

this liminal, crisis situation, the Bururi Tutsi-Hima faction saw an opportunity to cement their status. The state repression started immediately. That very night, Ntare V was assassinated

(Russell, 2015a:75). On 2nd May, a bulletin of La Voix de la Révolution condemned the ‘agents of imperialism’ and ‘traitors of the nation’ who had ‘tried to overthrow the republic regime’ and told its audience to ‘rest assured, without doubt, the Republic, the armed forces loyal to the Revolution of November 2, 1966 and its leader, Colonel Michel Micombero, will control the situation’ (in Chrétien and Dupaquier, 2007:151; see also Russell, 2015a:77). Micombero himself denounced the rebels, ‘driven by barbaric and inhuman sentiments’ (in Russel,

2015a:79) and claimed that there was evidence of a Rwandan-style Hutu plot to exterminate the Tutsi people.

By 30th May, whilst counterinsurgency operations were ongoing, segments of the armed forces and the JRR initiated efforts to exterminate all suspected Hutu rebels (Lemarchand,

1994:96-97). Massacres were then extended beyond the southern provinces across the country as a whole. The criteria for suspects was eventually extended to mean ‘educated’ Hutu people, a term which left room for interpretation – those with glasses were suspects, as were primary students, secondary school and university students, civil servants and political elites (Prunier,

1995a; Chrétien and Dupaquier, 2007:204-241). Memorialised as ikiza, meaning the ‘scourge’ or ‘catastrophe’ (Chrétien and Dupaquier, 2007:81), this was a veritable genocide.

In this sense, I depart from some qualifying and specifying terms that have been used in the scholarship on this event. Lemarchand and Martin (1974) referred to ikiza as a ‘selective genocide’ on account of the specificity of the victims’ education and social status within the category of Hutu identity. Relatedly, the term ‘eliticide’ has also been coined and applied to

Burundi (e.g. Jones, 2011:26). I call it genocide for three reasons: firstly, because the criteria for defining the Hutu elite was very broad and inconsistently applied; secondly, in agreement with Shaw (2007:64), terms like ‘eliticide’, ‘politicide’ and ‘ethnocide’ each encapsulate 232

differing elements of the more overarching concept of genocide rather than phenomena distinct from genocide; and thirdly, because such terms reduce genocide solely to mass killing. As we shall see, genocide was here employed in order to change the relationship between perpetrators and survivors (Feierstein, 2014). Again with Shaw (2007:34), I suggest that ‘genocide involves mass killing but it is much more than mass killing’.

The armed forces and paramilitary groups were well organised and had an arsenal of relatively sophisticated weapons at hand, including machine guns, armoured vehicles, grenades and even helicopters (Greenland, 1976:120; Abrams, 1995:152; Chrétien and Dupaquier,

2007:193, 197). In many ways, this resembled the 1965 coup and repression, which had been a state-led exercise against political elites. However, much more of the population were implicated here, as perpetrators as well as victims. In a development evocative of RTLM in

Rwanda during the 1990s, messages were broadcast on government radio that encouraged Tutsi people in the countryside to ‘hunt down the python in the grass’ (Greenland, 1976:120). As with civilian participation in Rwanda, some obliged out of fear and others out of personal grievance or material calculations (Lemarchand, 1994:102; Chrétien and Dupaquier, 2007:194).

A number of Rwandan Tutsi refugees, living in exile in Burundi after their own traumatic experiences of ethnic violence, participated in massacres (Lemarchand, 2009:139). Many people were taken away to be killed in isolated spaces, often with obedience and placidity

(Lemarchand, 1994:98). Such a response was testament to the methods and total organisation of the genocide (Russell, 2015:450-451), as well as to traditional legacies of adherence to authority94 certainly present in the precolonial period and intensified under colonial rule (for discussion see Chrétien and Dupaquier, 2007:188-189).

94 Michel Kayoya, in 1971, referred to a ‘conscience féodalisée’ to describe these attitudes of resignation (Chrétien and Dupaquier, 2007:189). A priest, poet and intellectual, Kayoya was himself killed in the genocide on 13th May 1972. 233

There are several reasons that ikiza has been ‘forgotten’ (Lemarchand, 2011) or ‘hidden’

(Hinton et al, 2013) – at least outside of the Great Lakes region – in the years following its occurrence. The events went largely unreported at the time, owing to international apathy and indifference in addition to the tight control that the Micombero government wielded over international institutions and journalists (see Lemarchand, 1980:60; Chrétien and Dupaquier,

2007:363-464). What is more, they occurred at a time when commitments to atrocity prevention and human rights were still in their infancy and constrained by their entanglement in the politics of the Cold War (Therborn, 2011:1).

Nevertheless, a large factor in its slide into oblivion has to do with its grim success in instrumental terms. It is profoundly disquieting to think that a genocidal operation might come to an end based not on external intervention or an internal counter-force, but rather as a result of its accomplishment in the eyes of the perpetrators, all the more so when those perpetrators largely had their social status cemented or upgraded in the years of impunity afterwards (see

De Waal and Conley-Zilkic, 2006). ‘In area after area’, according to a witnessing US embassy official called Michael Hoyt (in Lemarchand, 1994:103), ‘no educated male Hutu is believed to be alive. This is particularly true in the south where we have word from growing number of villages that no Hutu males remain at all’. Hutu people were purged from all areas of public and civil administration, from politics and education. As Lemarchand (1994:103) details:

For the next fifteen years, only Tutsi were qualified to gain access to power, influence, and wealth. Through the transfer of the victims’ property (and relief funds) into Tutsi hands, one of the more bizarre consequences of the 1972 bloodshed has been to reinforce significantly the position of educated Tutsi as a dominant class. Not only access to the state but also participation in violence thus became a key ingredient of class formation. To an even greater extent than before, what was left of Hutu society was systematically excluded from the army, the civil service, and the university. Never before had the Hutu as a group been so thoroughly reduced to the status of an underclass. From all appearances, only Tutsi were fit to rule, and among them, none were presumably better qualified than the ‘Banyabururi’ – the people from Bururi… More than anything else, Hutus’ shared consciousness of being a martyred community gives them their sense of forming a group apart. Their concrete, everyday experience of subordination and oppression serves only to reinforce their awareness of being the expiatory victims of Tutsi hegemony. 234

SECOND REPUBLIC Tutsi-Hima hegemony was assured in the aftermath of the genocide but the Micombero government emerged in disarray, dominated by military figures and JRR militants

(Lemarchand, 1994:107). The Jacobin commitments of the 1960s had dissipated, and in terms of an animating ideology there remained only vague references to the ideas of ‘progressive nationalism’ common in Cold War-era Africa (Chrétien and Dupaquier, 2007:51). Micombero himself was said to have retreated into drink and paranoia (Prunier, 1995a:11).

On 1st November 1976, Jean-Baptiste Bagaza, Micombero’s cousin, declared a coup and established a new republic. The unifying theme in the discourse of the new regime was

‘national unity’:

The message, relayed by the media, came through clear and loud: nowhere is the ancestral unity of the Barundi more evident than in their common language (hence, the importance of making Kirundi the only medium of instruction in primary and secondary schools), in the social solidarity of the peasant masses (hence, the significance of rural cooperatives and ‘villagization’), and in the rich body of oral traditions that once formed the social cement of the nation. Only through a revolutionary projet de société could this rich heritage be restored to its proper place and the contradictions of the First Republic eliminated. ‘The fundamental contradiction which nearly destroyed the nation’, in the words of the Charter of Uprona, ‘is between the Barundi people and a reactionary bourgeoisie, which stands as the principal enemy of our struggle’ (Lemarchand, 1994:107). References to ethnic identity were banned and the idea of unity was conceived on revivalist lines. A programme of ‘Kirundisation’, similar to Mobutu Sese Seko’s programme of ‘Zaïreanisation’, was initiated. Comprised of an attempt to make Kirundi the lone language taught in schools, it left French speaking to an urban class of Tutsi elites (Lemarchand,

1994:109). The programme was also made manifest in a renewed concern with the national cultural traditions of Burundi. As Emile Mworoha (1986:354; see also Mworoha, 1987), an intellectual architect of the regime, put it:

It is now known that no valid development project can actually succeed if it does not seek to integrate cultural elements in the population concerned. This was understood by the Second Republic, which since its beginnings has created state cultural services, previously unknown, and has involved itself in the activities of cultural dissemination, research of the oral traditions 235

of the country, the organisation of the national archives and the preservation of cultural heritage in general.

In rural areas, the regime instigated a policy of villagisation. This was unpopular as it was perceived, with large justification, to be a mode of increasing regulation and surveillance over the rural population. It operated by way of regrouping disparate rural populations into villages, on the assumption that larger units of production would increase agricultural yields.

Life in the countryside was regulated dramatically (Lemarchand, 1994:108-110; Prunier,

1995a:11) and was accompanied by an ideological conflict with the Church, defined by

Mworoha as part of a programme of ‘liquidating the last vestiges of colonialism’ (Brooke,

1987:12; see also Lemarchand, 1994:112). This can be tied to Tutsi suspicion of the Catholic

Church for their role in the Rwandan revolution. Their perceived role in Hutu education was viewed with suspicion, and many initiatives were shut down and seminaries nationalised. This was aided by fear on the part of Hutu families of sending their children to school, caused by the killings of educated Hutu in 1972 (Kay, 1987:7; Malkki, 1995:36-37; Skonhoft, 2000).

At the beginning of the 1980s, there emerged a number of political movements that advocated the return of refugees. These refugees were based in camps located in neighbouring countries like Tanzania, Rwanda and DR Congo. The groups formed associations that later became political parties (Rutake and Gahama, 1998:95). Mirroring the formation of the RPF in southern Uganda following the Rwandan revolution of 1959-61 – comprised, as we have seen, of Rwandan Tutsi exiles – there emerged the Parti pour la libération du peuple Hutu

(PALIPEHUTU), principally in the Mishamo camp in Tanzania in 1980. It was created by Hutu refugees exiled by the 1972 violence and led by Rémi Gahutu. The camp refugees saw themselves as ‘a nation in exile, and defined exile, in turn, as a moral trajectory of trials and tribulations that would ultimately empower them to reclaim (or create anew) the ‘homeland’ in

Burundi’ (Malkki, 1995:3). These future-orientations are the basis of what Liisa Malkki (1995) calls ‘mythico-histories’, which were: 236

Concerned with the ordering and reordering of social and political categories, with the defining of self in distinction to other, with good and evil. It was most centrally concerned with the reconstitution of a moral order of the world. It seized historical events, processes, and relationships, and reinterpreted them within a deeply moral scheme of good and evil (Malkki, 1995:55-56). Biographies were subsumed into this collective narrative of cultural trauma

(Alexander, 2016); individuals were ‘thrown together’ in exile and developed a strong collective consciousness because of their Hutuness (Malkki, 1995:102; see also Christensen,

1985; Turner, 1998). Importantly, before 1972, one could not speak of a Hutu ethnic consciousness among the masses based outside the political and intellectual scene in

Bujumbura. In distinction to the Rwandan Tutsi exiles of the RPF, who were united in spite of other differences by a shared sense of Tutsiness resulting from the ethnopolitical conflict of the

1950s and 1960s, the Hutu population of Burundi neither felt that they constituted a political entity, nor saw their status and role in ethnic terms (Chrétien and Dupaquier, 2007:191). What produced ethnic identity on a mass scale, and became inextricably tied to an interpretation of ethnonational belonging, was the collective trauma of the experience of genocidal violence. As an informant in Malkki’s (1995:101) study said: ‘What made us think about our country was the troubles of ’72. The troubles of ’72 awakened us because the Tutsi killed the children, the old people, the pupils in schools, the pregnant women and… all the inhabitants of the country of Hutu origins’.

THIRD REPUBLIC 1987 saw another coup, again a result of incremental regime degradation but in particular provoked by the decision of Bagaza to force the retirement of a number of military officials in a context of austerity (Lemarchand, 1994:116). A year later, there emerged an uprising of Hutu peasants, co-ordinated by PALIPEHUTU, in two rural communes along the border with Rwanda (Lemarchand, 1994:122). Ethnic animosity here was particularly strong because it was where large numbers of Rwandan Tutsi exiles resided, possessing memories of 237

Hutu-led persecution in their country of origin. Further still, it was in this region that the global coffee market crisis that had also hit Rwanda had its most potent effect in Burundi, combining with competition for scant social services. Local Tutsi officials also made repeated provocations, which in turn made a violent response seem more and more likely (Lemarchand,

1994:122-124).

Fear of a re-run of 1972 also played a large role in prompting an uprising (Lemarchand,

1994:127). As the PALIPEHUTU cadres – the organisers of this affair – recommended in their tracts, one had to ‘surpass the Tutsi’ and ‘do the work’ necessary to neutralise the ‘project of

Hutu extermination’ (Chrétien, 2003:319). President Buyoya was accused of planning this project, and he was compared to Michel Micombero. As happened previously, the scenario of provocation and repression played itself out across the country. Thousands perished, more than

50,000 Hutu fled to Rwanda, and the international community denounced Bujumbura’s Tutsi regime (Loft, 1988; Lemarchand, 1994:126; Chrétien, 2003:319-320).

However, the regime after this event tried to find a way out of the ethno-political crisis.

They attempted, still under the stewardship of Pierre Buyoya, to forge a government based on power-sharing between ethnic groups, and introduced mechanisms to condemn exclusion and discrimination. In spite of this, there was still great tension; there were massacres in 1991 as well as repressions of Hutu (Lemarchand, 1994:152). Under the international pressure of the

Arusha Accords, as in Rwanda, Burundi adopted a democratic constitution in 1992, including provisions for multiple parties (Boyer, 1992; Lemarchand, 1994:131-132; Chrétien, 2003:319-

320). A party called the Front pour la Démocratie au Burundi (FRODEBU), led by Melchior

Ndadaye, was founded as a mediating institution between the ethnic exclusivism of

PALIPEHUTU and the regime-supporting UPRONA. In June 1993, elections were held and

Ndadaye was made head of state, becoming the first civilian and elected president of the republic. As Chrétien (2003:320) states: 238

This election, complemented by FRODEBU’s success in legislative elections, was perceived on all sides as Hutu revenge. But the new president affirmed his desire to eradicate the ‘ethnic sickness’, and he formed a government that was a third Tutsi and presided over by a technocrat close to UPRONA, Sylvie Kinigi. Despite evident tensions on both sides, the country seemed to be moving beyond its tragic destiny. Not long after, on 21st October 1993, Ndadaye was killed by Tutsi-Hima insurgents based within the military. Prunier (1995a:26; see also Nahimana and Hatungimana, 2004) recounts the tragic extinguishing of hope following the death of Ndadaye, who seemed to be matching rhetoric about unity and democracy with action:

President Ndadaye was not only a Head of State, he was an almost Christ-like figure who had come to symbolically release his people: from bondage. This feeling was due to history and symbolical politics, but it was also reinforced by the deepening economic crisis into which Burundi, like the other countries of the area, was gradually sinking after world coffee prices had started a rapid decline in 1987-1988. The assassination of Ndadaye led, somewhat predictably, to a perception that a Tutsi-Hima repression was soon to stem the democratic tide. Hutu bands formed in the hills of rural areas and massacred Tutsi, justified as ‘self defence’, and the largely Tutsi-Hima military responded in kind. This was the beginning of over a decade of civil war, estimated to claim the lives of around 300,000 people. This event also had regional repercussions, significantly radicalising extremist Hutu factions in Rwanda and undermining the power-sharing agreements of the

Arusha Accords (Des Forges, 1999:101).

INTERMEDIATE REFLECTIONS

I now want to expound some comparative statements about the histories just detailed, specifically in relation to the overall aims of this thesis and its original contributions. In the first instance, I argue that the violence of various scales, including the genocides of 1972 in

Burundi and 1994 in Rwanda, conforms to what I earlier suggested might be termed the

‘creativity of intentionality’ 95 (see Joas, 1996; Schneiderhan, 2013). Genocide has tended overwhelmingly to occur as a situational form of action in the context of economic and political

95 See pages 27-34. 239

crises that are emergent from and exacerbated by exogenous factors. Experiences of acute societal crisis, when social life ceases to appear as a ‘reality without question’ (Cordero,

2017:1-2), are the conditions for interpretation par excellence. These situations have the potential to guide the routes to and through modernity in novel and unprecedented directions.

Genocide represents an extremely destructive and purposive, if improvisatory, attempt to close the space opened up by crisis situations. The extremely violent reconfiguration of social relations (Feierstein, 2014) is thus an important facet of modern genocide.

In this sense, it is telling that it is very often the case that the first people to be killed in genocide are political moderates and opponents, whose critique might offer an alternative. The names on the lists of the Rwandan National Guard to be extinguished in the period after

President Habyarimana’s death were moderates and members of political opposition groups.

Similarly, in Burundi in 1972, educated Hutu groups were the principal targets. In both instances, the architects of genocide came from powerful elite groups who, paradoxically, possessed powerful self-perceptions of disenfranchisement and marginalisation; in Burundi, these have historically been the Tutsi-Hima who comprise the Bururi lobby, while in Rwanda they have been Hutu people of Bakiga lineage based in the northern regions of the country.

Genocide has occurred when these groups feel their status and security to be under threat. The post-independence violence of Burundi was situated within the punctuations of interfactional coups, successful and unsuccessful, in the urban setting of Bujumbura; Hutu insurrections were located in the rural provinces and across the borders. In Rwanda, anti-Tutsi violence is inextricably tied to the insurgencies of armed groups linked to those exiled by the violence of the 1959 revolution. As scholars of genocide have noted (Shaw, 2002:95-96; Moses,

2008; Travis, 2016), genocide has an elective affinity with state policies of counter-insurgency.

Counter-insurgencies may not always be genocidal, but genocide and ethnic conflict is almost 240

invariably sanctioned in the name of national security, law and order or some such similar vocabulary, whether or not the reality of the threat matches the discourse.

Another original contribution presented here, which is tied to the notion of the

‘creativity of intentionality’ and much implicates genocide as a practice related to modernity, is the temporality of genocide. In both Burundi and Rwanda, though in differing idioms, genocide has been committed with a ‘future-orientation’. In Burundi, in 1972, terror was intended to transmit intergenerationally. The aims of the perpetrators were to ‘spread terror throughout the entire community and thus create an enduring sense of fear and submission among the living and the unborn – in short, to teach a lesson that would be remembered by generations to come’ (Lemarchand, 1994:102), and to ‘lay the foundation for an entirely new social order’ (Lemarchand and Martin, 1974:15; Abrams, 1995:148).

Feierstein’s (2014) theorisation of genocide as a social practice is relevant here.

Genocide was employed to reconfigure social relations in such a way that the privileges and power of an elite group would be secured. The Burundian genocide is an example of a successful bout of genocidal violence in this regard, the more so for its fall into oblivion both in official discourse in Burundi and in the outside world. Lemarchand and Martin (1974; see also Kay, 1987:5-7) express amazement at how the once ‘relatively open and flexible system of stratification’ that the country had known since precolonial times had been transformed into a society where access to material wealth, education, status, and power’ were restricted to representatives of a ‘dominant minority’ of Tutsi-Hima from Bururi. Violence, in this sense, was aimed at those who remained as much as at those who were killed.

In Rwanda, the future-orientation was tied to the memory of the foundational moment of revolution. As Larry Ray (1999:15) has written: 241

Every modernist revolution is ‘unfinished’ because its project is impossible. Thus, like the Jacobins, they take on a prophetic quality that permeates later political consciousness. Modernist revolutions, like Janus, faced in two directions – towards a past of community, solidarity and authenticity; towards a future purged of alienation and doubt that nonetheless sustains modernity’s technical and organizational potentials. The social revolution idea was deployed frequently throughout the civil war period. Tutsi people were denounced principally in pre-genocide propaganda as monarchists and feudalists.

Professor Anastese Gasana, a technical adviser to the MRND leadership, said during an interview in 1990 that the RPF insurgency was launched in order to re-establish ‘a monarchic and feudal regime that certain Rwanda refugees want to have restored in Rwanda… the

Rwandan refugees never acknowledge the 1959 revolution’ (Radio Rwanda, 1990). As Des

Forges (1999:195) puts it:

The ‘intellectuals’ of Butare discussed the need for ‘uniformity and harmony’ of language at two meetings that they held during the genocide. In official statements made at meetings, in correspondence among administrators and politicians, and in radio broadcasts, this ‘uniformity and harmony’ prevailed and in the vocabulary used even long after the fact by participants, it still prevails. Some ordinary words carry a special meaning, like ‘to work’, which appears frequently and almost casually, meaning to kill Tutsi and their Hutu supporters. The word refers back to the 1959 revolution and its violence against Tutsi, a link indicated in phrases that advocate ‘finishing the work of the revolution’. ‘Work’ requires ‘tools’, that is, firearms, machetes, clubs, spears. This tied into the specifically agricultural interpretation of development. ‘Work’ had connotations of the noble toil of the autochthonous peasantry. Killing was related to practices of collective community development work (umaganda), and was euphemistically referred to variously as ‘tree felling’, ‘bush clearing’, and in the case of children and the women from whom they were born, as ‘pulling out the roots of the bad weeds’ (Keane, 1995:165; Prunier,

1995:138-142; Uvin, 1999; Hintjens, 2001:39).

The idea of Rwanda as a proletarian Hutu nation and the animating societal self- understandings of ethno-republicanism resonated with the practical, material concerns of Hutu people across the country, particularly those related to land:

As a nation of farmers in a country short on land, Rwandans had been concerned about control over property for many years. Anti-Tutsi propagandists exploited the issue even before the 242

genocide began by suggesting that the RPF intended to overthrow the 1959 social revolution and repossess the lands that Hutu had acquired after killing Tutsi or driving them out of their communities in the 1960s (Des Forges, 1999:433; see also Lemarchand, 1995:8-9). Ethnicity is instrumentalised by state elites, rather than being a primordial entity, but it requires precipitating conditions for it to be instrumentalised, including a receptive population.

Furthermore, it is not enough to identity certain endogenous conditions. Internal political crises are themselves tied to certain precipitating factors which are exogenous. At the level of societal entanglement, this includes the tumultuous and transformative international situation of the Great Lakes region. We have seen that Burundi and Rwanda’s post- independence histories were entangled in ways that had important repercussions in terms of violence: for their fear-mongering about the return of Tutsi ‘feudalism’, the advocates of Hutu

Power in Rwanda drew on the examples in Burundi; for their fear-mongering about the elimination of Burundi’s Tutsi population at the hands of a tyrannical majority, the Bururi faction drew on the examples in Rwanda, beginning with the 1959 revolution.

Furthermore, refugee camps proved fertile grounds for the emergence of restitutionist political movements, such as the RPF and PALIPEHUTU, whose aim was to ‘return home’ following an experience of violence and expulsion. In these borderlands, they became embroiled in the political struggles of the host-country; the RPF had been discriminated against by the Milton Obote government in Uganda, and helped Museveni to power. In Burundi,

Mulelists expelled by the Zaïrean government of Mobutu Sese Seko were found among refugees and were implicated in the rebellion in 1972.

Geohistorical entanglements were also important. The initial bout of violence in

Burundi in 1965 had been preceded the year earlier by a devaluation of the Burundian franc

(Rutake and Gahama, 1998:79). However, it is important not to talk about the weakening of the state or the prompting of its failure in any uniform sense. Genocide relied on the strength 243

of the state. Andy Storey (2001; see also Uvin, 1998:59) argues that the World Bank and the

IMF strengthened state power, rather than weakening it. There is a parallel to the colonial state here. Rather than a blanket and uniform degradation and dissolving of power, we actually see a strengthening of its elites and their capacity to control the population. The elites, especially in Rwanda, used development discourses both to manage relationships with international institutions of humanitarian and development assistance and to entrench control over the domestic population (Uvin, 1997:100). This falls generally in line with the neoliberal policies and structural adjustments of the World Bank and the IMF, whose overall aims were a weak state in relation to international capital and a strong state in relation to labour (Hallward,

2001:132; Harvey, 2005).

There is also a matter related to the previous chapter, namely the difficulty of positing and instituting autonomous trajectories in a vastly unequal capitalist world-economy, an inequality rooted in the relationships of dependency established in and maintained following the colonial period (Rodney, 1972; Wallerstein, 1974; Jonsson and Hansen, 2014). These geohistorical entanglements bring with them external pressures – democratisation in the wake of the fall of the Berlin Wall, development assistance and the arrival of NGOs, structural adjustment policies implemented by the World Bank, global economic crises, etc. – which have significant impacts on the ground. The ‘dark side’ of democracy (Mann, 2004) in Burundi and

Rwanda was revealed under the auspices of international organisations such as the United

Nations and powerful states such as the USA, committed to democratising such places in the context of a post-Cold War world, as demonstrated by the Arusha Accords of 1992.

In sum, these geohistorical entanglements have severely limited any attempts to institute autonomous interpretations of modernity outside elite power groups. The ramifications of this are deep, from the capture of development funds by a ‘state class’ to the humiliation of grinding poverty for masses of people in the countryside. ‘The horizon open to 244

the future, which is determined by the expectations of the present’, said Habermas (1987a:13),

‘guides our access to the past’. Because the horizon open to the future is so constrained by these pressures, and the expectations of the present are subsequently disappointed, the past is accessed in a particularly violent manner. Furthermore, as we have seen, this past is conceived of in a framework that was racialised and traditionalised during the colonial encounter. It is in this violent confrontation between modern reconstruction on the one hand and racialised traditions and their historical narration on the other that we ought to locate the sort of ‘traumatic modernity’ hinted at by Eisenstadt (2000:25-26).

The modernity of genocide in Burundi and Rwanda is bound up with the disappointment of utopian aspirations and the attempt to reinvigorate them in dialogue with significant historical experiences, such as the 1959 revolution in Rwanda. Modern genocide is also derived from the exclusivist, elitist clamouring for resources and power, and inextricably linked to elite attempts to re-engineer social relations, as is particularly the case in Burundi.

Broader populations are brought into violence as perpetrators in no small part by the trauma, humiliation and shame of historical suffering, in combination with resource and land scarcity

(Lindner, 2009; Ray, 2013).

Thus, the Burundian and Rwandan cases are both expressions of modern genocide.

They both relied on strong state bureaucratic administrations and on the massive collective mobilisation of the population. Furthermore, in both cases genocide was expressed in a reconfigurative, future-oriented sense, inasmuch as genocide was employed as a means to realise elite ambitions and transform social relations (Bauman, 1989; Feierstein, 2014). Genocide was also employed as a means to the end of mastering social and political crises, which were themselves resultant from the dynamics of global modernity. In short, it might be said that episodes of genocide were extremely violent responses to the crises and disappointments, even 245

the humiliations, of postcolonial modernity; of the unrealised and frustrated possibilities of independence.

246

PART THREE

CHAPTER NINE – CONCLUSION

In this final section, I want to draw some conclusions from this study and point to some directions in which further work might proceed. To begin with, I summarise the original contributions that the thesis makes to debates in social theory concerning the multiplicity of modernity, principally at the intersection of the paradigm of multiple modernities and postcolonial sociology. I then outline how various aspects of the thesis point towards work that can be addressed in a future research agenda. In a final section, as is consistent with the hermeneutic thrust of this project, I return from the ‘exterior’ historical subject of the thesis to more ‘interior’ matters in the here and now and make some brief, more value-oriented claims concerning present-day problems and challenges. Firstly, I consider the conclusions in relation to the contemporary socio-political scene of the Great Lakes region, with a particular focus on

Burundi and Rwanda. Secondly, I assess some developments in contemporary Western societies.

ORIGINAL CONTRIBUTIONS

The original contributions of this thesis are delineated in three sections. Firstly, I summarise the substantive contribution of the thesis. The preceding analysis of the entangled histories of Burundi and Rwanda is presented as a new case to the study of historically and geographically differentiated experiences and interpretations of modernity. This case necessitates confronting and incorporating the themes of indirect colonial rule and genocidal violence, which are neglected themes in the literature hitherto. Secondly, I turn to the theoretical contribution. In considering these experiences of colonial-imperialism and genocide, I identify some areas in the ‘multiple modernities’ paradigm, as well as in postcolonial sociology, which require extension so that they are more encompassing of a 247

variety of trajectories and experiences of modernity and their ‘entanglements’. Finally, I highlight two secondary original contributions that the thesis makes by virtue of the subject matter that it addresses. The first of these is that it contributes a novel conceptualisation of the relationship between modernity and genocide. In addition, with its emphasis on philosophical hermeneutics, the project presents an original contribution to debates about Eurocentrism in the discipline of sociology, its operational concepts, and its ability to address the historical experiences of non-Western societies.

SUBSTANTIVE CONTRIBUTION Taking my point of departure from Göran Therborn’s (1995, 2003) suggestion that the experience of colonialism constitutes a distinct route to and through modernity, one which is entangled with the European route to and through modernity, I argued that the experiences of indirect rule in Burundi and Rwanda are at the same time specific experiences of modernity.

The effects of this specific articulation of modernity, however, are paradoxical. Indirect colonial rule, as a form of colonial modernity more broadly, possessed a modern future orientation in the sense of its intentions to modify entire social structures in order to bring

‘progress’, ‘development’ and ‘civilisation’ to indigenous populations and to open up local economies to global capital. This utilitarian orientation, though, existed alongside the traditionalising and racialising frameworks used by colonialists to understand and order these societies. Colonial modernity in its indirect form, as experienced in Ruanda-Urundi, was marked by a fundamental tension between the instrumental-rational motivations towards making the population productive, on the one hand, and the romantic valorisation and preservation of difference as encapsulated in the Hamitic Hypothesis, on the other. This was expressed especially in the reconstruction of institutional structures of society along racial grounds, influenced by the Hamitic Hypothesis. 248

In this crude racial schema, Tutsi people were Hamitic conquerors and born rulers and

Hutu people were subservient agriculturalists. A number of important institutional reforms were enacted on this basis. Burundi was deemed to be a variant of Rwanda in its societal constitution and was subject to the same reforms, in spite of its different precolonial history.

As such, a contradiction arose: colonialism introduced new demands and needs among the population, and also fatally undermined some precolonial institutions with which it was incompatible. At the same time, the racialisation of tradition solidified social relations and social conflicts. To be administered racially is to be ‘fixed’ or ordered in time and space.

Independence movements emerged in the 1950s, and in the tenor of elite social discourses we can observe strong commitments to the modern imaginary significations of autonomy, social justice, progress, and so on. Although these arose especially in the communications of elite groups educated in Catholic seminaries, they cannot be explained away as the emanations of puppets or dupes of Western authorities. The political parties formed on the basis of these interpretations had widespread public support. Presented in their discourses are distinctly modern societal self-understandings, open towards an autonomous and shapeable future. At the same time, these imaginary significations are interpreted against the background of experiences of colonial injustice and with reference to the historical horizons of the traditions that had been racialised under indirect colonial rule.

Moreover, these societal self-understandings were articulated in a material context derived from a confluence of structures and processes effective over and above the level of the nation state. Variously, Cold War rivalries, the legacy of conflict in Europe, economic and legal relations of dependency with international institutions and powerful nations, and changing attitudes among Catholic missionaries each had important influences on the conditions for independence movements in Burundi and Rwanda, and on their specific directions. 249

At the same time, precolonial dynamics were also important for variations in routes to independence, especially the differences in the formation of monarchical states in Burundi and

Rwanda outlined in chapter five. On the eve of independence, conflict emerged on faultlines which preceded European colonialism, seen among ganwa in Burundi and in republican anti-

Nyiginya tendencies in Rwanda. These were certainly modified by colonialists in racial terms, but they were not invented by them ex nihilo. In Burundi, discourses of national unity, before the osmosis effect of the Rwandan revolution, were successful because of the weak institutional role of the monarchy in precolonial developments and their undermining by successive

European administrations. Independence programmes in Burundi were more straightforwardly anti-colonial and mobilised around the symbols of kingship. The Hutu-Tutsi distinction was mobilised in Burundi principally by postcolonial elites, with significant influence drawn from events in neighbouring Rwanda. In Rwanda, the Jacobin language and event of social revolution triumphed because ‘traditional’ monarchical Rwanda, embodied in the Nyiginya

Kingdom, was identified with historical injustice by a republican Hutu counter-elite. This animating idea played a crucial role in the construction of a majoritarian Hutu collective identity, mobilised in times of postcolonial crisis.

The postcolonial experiences of Burundi and Rwanda, exemplified in respective episodes of genocide in 1972 and 1994, are examples of what Eisenstadt (2000:25) called

‘traumas of modernity’, which result from the tension and dialogue between modern reconstruction and seemingly traditional forces. The seemingly traditional forces can be located in the racialised traditional framework introduced in the colonial period. The modern reconstruction can be found in the competing discourses and societal self-understandings of the decolonial period and their legacy following formal independence. In the postcolonial period, these independence discourses and competing interpretations of the modern social 250

imaginary were increasingly constrained by the reality of a marginalised position in a condition of global interdependence.

In this condition, the discourses of national unity in Burundi and of social revolution and development in Rwanda were captured by state elites and attached to a genocidal politics that emerged during times of economic, political or military crisis. In Feierstein’s (2014) terms, genocidal social practices were aimed at the mastery of crises and the reorganisation of social relations to secure access to power in the long term, in both cases on the part of particular minority groups – Tutsi-Hima from Bururi in Burundi and Bakiga-Hutu from the northern provinces in Rwanda. Historical experiences of violence also fed into the narrative constitution of collective identities, which had a pivotal role in enabling the sorts of mass collective mobilisations of society required for further periods of genocidal violence.

THEORETICAL CONTRIBUTION I have suggested that modernity is best conceived as a time-orientation, open to a future conceived of as a space for the projection of the possibilities within the present. These possibilities are situated within a field of tensions and require the active and often conflictual interpretation of modern imaginary significations. These interpretations are formed on the background of distinct historical experiences (of colonialism, of revolution or of war) and involve an active engagement with socio-cultural traditions. These interpretations unfold processually along specific historical trajectories or routes and proliferate in a variety of specific, institutional forms. Ethnocentric assumptions about the universality of the Western model of modernity are therefore untenable. Furthermore, distinct articulations of modernity do not proceed in a condition of separation. Instead, they are entangled both with one another and with the structural problems and transformations of international and transnational relations of power. 251

I have suggested that the framework of ‘entanglement’ serves to correct some problems within existing conceptions of multiplicity and diversity in terms of experiences of modernity.

I argue that entanglements can be delineated along three separate lines. Firstly, there are the entanglements of modernity and tradition constitutive of any given formation of modernity. An original contribution of this thesis is the argument that the colonial route to and through modernity, particularly in its indirect form, produces a distinct variation of this entanglement.

Here, tradition is racialised and thus certain customary forms of authority and governance that are seen to be compatible with the order-building imperatives of colonial administrations are solidified and fixed along racial lines.

This challenges some common assumptions about the relationship between modernity and tradition. The substantive analysis presented provides additional support to arguments against the idea prevalent in modernisation theory, carried on in notions of

‘detraditionalisation’, and inverted in certain strands of decolonial and postcolonial theory, that modernity and tradition are binary entities loaded with normative connotations. The experience of modernity introduced by indirect rule is not one of a progressive ‘detraditionalisation’, or indeed a steamroller-like flattening of indigenous customs and traditions.

This corroborates ideas presented by theorists of multiple modernities concerning the preservation of tradition in any formation of modernity. Nevertheless, notions of multiple modernities are also challenged: tradition in the colonial route to and through modernity was racialised, and the solidification of social relations along racial lines benefitted precolonial power elites with whom colonialists collaborated. The ‘cultural programmes’ (Eisenstadt, 2000,

2001, 2003) of Burundi and Rwanda were themselves colonised. Precolonial social conflicts and grievances were hardened and fixed in time. 252

Tradition, then, is not a stable and value-neutral repository that societies tout court access and turn toward an open future. As we saw especially in the analysis of independence movements presented in chapter seven, conflicts over relationships to traditions – relationships solidified in racial form in the experience of colonial-imperialism – themselves shaped the responses to colonial rule and the emergence of autonomous interpretations of modernity. It is in this context that genocide in Burundi and Rwanda has to be situated. Modern genocide is not to be derived from a lack of Western institutions, nor as a pathological variant of them, but rather should be seen as a product of the dialogue between racialised ‘traditional’ sources and modern reconstruction emergent from the specific routes to and through modernity undertaken by Burundi and Rwanda.

The second type of entanglement is that between different societies, which I have termed societal entanglement. Most obviously, Burundi and Rwanda can be seen as entangled in this sense. Their precolonial histories were connected along such lines as trade, ecology, migration and war, but they also diverged in a number of significant ways. Their entanglement was institutionalised during their experience of colonial rule, whereupon they were administered collectively as Ruanda-Urundi under Belgian rule. Despite its precolonial differences from Rwanda, Burundi was administered as if it shared its ‘traditional’ social structure. Ruanda-Urundi itself was attached to Belgian Congo. For some scholars, such as

Joseph Gahama (2001), Ruanda-Urundi was effectively administered as a sub-colony of

Belgian Congo. Colonised societies and colonial administrations are entangled with one another and this has significant economic, demographic and social consequences for the specific shape that colonial modernity takes. Moreover, these entanglements have significant effects in the post-independence period relating to the mimetic events of violence in Burundi and Rwanda. 253

As such, an original contribution offered by this thesis is a demonstration of the entangled histories of Burundi and Rwanda. This contribution is principally oriented torwards the field of genocide studies, which has isolated Rwanda from its entanglements in the Great

Lakes region. Resultedly, the Rwandan genocide in 1994 is included in a canon of cases alongside the Holocaust, Armenia, Cambodia and former Yugoslavia (Bloxham and Moses,

2010:4) while similar events in Burundi have often been ‘forgotten’ (Lemarchand, 2011) or

‘hidden’ (Hinton et al, 2013). The historical societal entanglements of Burundi and Rwanda demonstrated by the analysis presented in this thesis highlight the problems and, ultimately, the untenability of such ‘distancing’ of genocides from their regional and international contexts.

Consequently, the thesis complements the call made by Shaw (2013:63) for genocide studies scholars to work towards the development of an ‘international relations of genocide’ that ought to place:

…questions of historical transformation at the centre of analysis, and to formulate more precise theories of changing international and social contexts. If genocide is not a general feature of international relations in all periods and regions – even if it is not as rare as some think – the task is to identify particular temporal and geographical clusters of genocidal violence, the specific international conditions in which they are located, and the larger international dynamics involved. Any adequate ‘international’ theory of genocide must offer a complex understanding of the changing relations across and between societies, as well as between states, in the historical transformations of modernity. In this sense an international theory of genocide must be framed not sociologically, but in historical-sociological terms. Societal entanglements do not only pertain to genocide during the post-independence period. They also capture the shared colonial experience of Rwanda and Burundi as administered as Ruanda-Urundi. Ruanda-Urundi was also entangled with the colonising societies, firstly in weak form with Germany and then later with Belgium. We saw that the

Belgian administration governed with a particularly paternalistic style, supported by Catholic missionaries, under conditions of constraint from the metropolitan government as well as the administration in Belgian Congo. These sorts of entanglements also contribute to distinct forms 254

of colonial modernity. Colonial modernity is experienced in different ways that are partly dependent on cultural and institutional factors of colonial rule.

This presents an original contribution to an emerging postcolonial sociology. I have drawn much influence from its concern with the problem of Eurocentrism and its normative invocations towards an increased institutional reflexivity and relativisation of social science disciplines. Nevertheless, influential work within postcolonial sociology (Bhambra, 2007,

2014; Go, 2013, 2013a) tends to employ a homogenising notion of ‘European’ or ‘Western’ colonialism that flattens the specificities of colonial-imperial rule. If sociological theorisations of modernity have historically disavowed the influence of colonial relationships in the formation of the modern world order, it is necessary to confront and understand ‘the colonial situation’ (Balandier, 1966 [1951], 1970) in its diverse forms and locations.

Thirdly, there are the geohistorical entanglements within which distinct forms of modernity are situated. This includes the entanglements with institutions and processes of transformation on the global scene. As we have seen, the historical routes to and through modernity in Burundi and Rwanda have been tied up at various historical junctures with institutions like the League of Nations, United Nations, International Monetary Fund, blocs of nations formed during the Cold War, and others. The entanglement of societies in the structural problems of global transformations of modernity gives rise to a number of conflicting interpretations and responses. These interpretations occur under conditions of coloniality, in the form of economic, legal and political dependency which have outlasted the formal event of independence of ex-colonial nations. Autonomous conceptions of postcolonial modernity are curtailed by material structures of global legislation and exchange to which postcolonial governments have to adapt or react. 255

This thesis presents an original contribution in the sense that it is a corrective to a certain tendency within notions of multiple modernities to overemphasise separation and endogeneity, whereby nation states and their traditional bases of legitimation are implicitly conceived as

‘cultural containers that are coherent and bounded and reproduce themselves over time’, and form their own distinct modernities separate from those of others (Wagner, 2009:254). Colonial modernity provided resources and symbols of protest, which the colonised could use to overturn colonial rule and which were employed in the construction of collective identities and in postcolonial centre building. At the same time, it also solidified the lines of racial and ethnic division that became the principal lines of conflict between multiple blueprints of autonomous modernity. Furthermore, legacies of economic and political ‘coloniality’ curtailed the capacity to formulate autonomous answers and institutional responses to the economic and political problematiques (Wagner, 2008).

SECONDARY CONTRIBUTIONS The thesis also makes two ‘secondary’ contributions. They are secondary in the sense that they are necessarily entailed in, but less explicit than, the more purposive contributions outlined above.

Firstly, in unpacking Eisenstadt’s (2000:25-26) intriguing though undeveloped references to the ‘traumas of modernity’, the thesis presents a novel way of framing the relationship between modernity and genocide that avoids some of the Eurocentric and evolutionistic theorisations of the relationship that can be found in genocide studies. Modern genocide, I argue, ought not to be conceived as a pathology of industrial societies at an achieved state of modernisation or as deriving from a particular institutional formation, but ought instead to be situated as a product of legacies and experiences of specific, entangled routes to and through modernity. Furthermore, modern genocide possesses a ‘creative’ intentionality (Joas,

1996; Schneiderhan, 2013) as a form of action compatible with the future-orientation of 256

modernity, tends to be employed in ‘situations’ of crisis, and is aimed at the mastery of them.

Furthermore, the temporality of genocide entails distinct references to the past. Previous rounds of violence and humiliation are experiences drawn into the narrative formation of collective identities and interpretations of modern imaginary significations (Wagner, 2008), leading to significant potential for ethnic mobilisation in times of crisis.

Secondly, this project offers a modest original contribution to epistemological debates concerning sociology, globality and Eurocentrism. Drawing on philosophical hermeneutics, I argue that it is precisely through the engagement with the exterior object of non-European historical trajectories that we can move to an interior understanding of the pitfalls of sociological theorisations of modernity that unduly universalise from the particular European experience. This complements those programmatic calls for a greater dialogical and interdisciplinary engagement on the part of social theorists with the ‘untapped sources’ contained within area studies literatures (Arjomand, 2014), to which I have adhered throughout this thesis.

At the same time, it has to be recognised that theoretical resources within social science traditions are what enable such a critique to emerge in the first place by rendering other trajectories intelligible and open to question. Eurocentrism is transcended not through the deconstruction and abandonment of macro-theorisations of societal development, but rather in an open-ended and reflexive outward movement of extension. This is not equivalent to a naïve assumption of objectivity and neutrality or an unquestioned adherence to habit or tradition. The disentanglement of sociologists from Eurocentric conceptions of historical development is rather an active process involving a gradual and reflexive movement from centredness in space and time (Harrington, 2016:53). To disentangle requires a priori some acknowledgement of the nature of one’s entanglement, even if one’s disentanglement can only ever be ‘relatively absolute’, to use Kurt Wolff’s expression (2002:53). 257

Understanding is not a matter of jumping ‘out of our skins’, of subordinating another person to our own standards, or of embracing radical relativism. Rather, it involves the attempt to rise or ‘spiral’ ‘to a higher universality that overcomes not only our own particularity but also that of the other’ (Gadamer, 2004 [1975]:305 emphasis added). Bauman (1978:217) writes similarly:

Understanding, as it were, is not about ‘feeling the feelings’ and ‘thinking the thoughts’ of others, but about sharing in a form of life. Or, in the case of an encounter between hitherto alien forms – about constructing a form of life of a ‘higher order’, which will incorporate the previous two as its sub-forms. This form of life of the higher order will contain all those ‘contiguous points’ where the previous two become elements of each other’s situation. An engagement with the histories of Burundi and Rwanda and their entangled routes to and through global modernity is not equivalent to a colonial-imperial tendency of the social sciences to ‘speak for’ the Other. Nor does it consist of a relativistic recognition of absolute particularity. Rather, the recognition of the constitutive multiplicity of modernity, and an acknowledgement of the diversity of historical routes and experiences within it, enables an orientation to the universal ‘we’ of humanity. Though any invocation of this ‘we’ must be qualified with reference to the spatio-temporal horizons of its origins, and must in principle be open to scrutiny of its own fallibility and provisionality, it is nevertheless possible to invoke it.

FUTURE RESEARCH AGENDA

I will now propose some directions in which further work might proceed. In the first instance, a future research agenda might be oriented towards broadening the geographical scope applied here so that it is encompassing of the Great Lakes region as a whole, including the multi-ethnic societies of Democratic Republic of Congo and Uganda. Beyond this, however, the current project opens up the possibility of comparison with a range of other historical trajectories to and through modernity, principally those with traumatic experiences of genocidal conflict and colonial-imperialism, such as Indonesia, East Timor and Cambodia. 258

A project which considers these diverse trajectories would necessarily have to confront the varying practices and institutions of colonial empire-states. This would constitute a welcome amendment to ahistorical and homogenising notions of ‘European’ or ‘Western’ colonialism. I contend subsequently that a future research project could be based around the analyses of what might be called varieties of colonial-imperial modernity. As much as the nineteenth century was one of dominance by Western European colonial empire-states

(Osterhammel, 2014:xx), it was also a period of multiple ‘civilising missions’ and interpretations of the international standard of civilisation. In the first instance, this refers to national interpretations and variations within Europe itself.

More broadly, however, such a research programme would be geared towards an assessment of the importance of rival colonial-imperial projects in ‘alternative’ modernities

(Arnason, 2000) or forms of ‘reactive modernisation’ (Therborn, 2011:61-63) in world regions and civilisational constellations beyond Europe. Among the possible objects of comparison here are the Russian (Geyer, 1987; Kaczmarska, 2016), Ottoman (Deringil, 2003) and Japanese

(Myers and Peattie, 1984; Suzuki, 2009) cases in the nineteenth and twentieth centuries. In such a project, I suggest, the concept of multiple modernities within the research programme of civilisational analysis can be brought into a productive dialogue with the imperatives of postcolonial sociology (Smith, 2015).

An original contribution of this thesis is a conceptualisation of modern genocide that draws on and extends the reference to ‘traumas of modernity’ that Eisenstadt makes in his paradigmatic essay on multiple modernities (2000:25-26). I have used this to analyse practices of genocide in Burundi and Rwanda. I have also suggested, however, that the theorisation of modernity promulgated by figures such as Eisenstadt, Therborn and Wagner, inasmuch as it principally entails a future-orientation and a distinctive conception of human agency, can also illuminate value commitments, practices and instruments of genocide prevention during the 259

twentieth century. In genocide studies, and in social theoretic work more broadly, it is underemphasised that genocide is also a phenomenon of modernity in the sense that its conceptual development led to its institutionalisation in the Genocide Convention of 1948, which named a crime perpetrated against humanity in its entirety and was oriented towards a future of stability guided by a principle of ‘utopia as never again’, to borrow Bo Stråth’s

(2016:9-22) expression. A ‘conceptual history’ (Koselleck, 2004 [1979]) of genocide along these lines would be a welcome addition to the sociology of modernity.

This theoretical and historical work could underpin more substantive studies of what might provisionally be called ‘traumatic modernities’. By this, I mean something distinct from the notion of ‘traumas of modernity’, whereby the relationship between seemingly traditional forces and their dialogue with cultural and political programmes can manifest in intense genocidal social conflict. I am suggesting an attention to those cases in which, to use Wagner’s

(2008) terminology, an experience of trauma itself forms the reference point for distinct interpretations of modernity, which in turn give rise to transformations in value orientation and institutional formation. Post-genocide Rwanda is a fascinating case in this regard, and fruitful comparisons might be made with countries such as Israel and post-apartheid South Africa.

Overall, this would contribute to what Joas (2000:76) has identified as a key frontier in studies of diverse trajectories of modernity, namely ‘the theoretical reconstruction of the connections between historical events, the structure of individual and collective experience, the articulation and imaginary interpretation of such experiences, and the resulting cognitive and evaluative orientations and their institutional expression’.

IMPLICATIONS FOR CONTEMPORARY DEVELOPMENTS

As the quotation from Arendt presented as an epigraph to this thesis suggests, my primary purpose in this work has been to understand the phenomena under investigation. I have 260

been less concerned throughout with influencing or providing policy recommendations in the present day. Valuable those these are, I have tried to step back from the temptation to ‘retreat into the present’ (Elias, 1987), and from the imperative to prescription. There is an equal value, especially in light of contemporary institutional contraints96, in historical sociology as a way of ascertaining the specificity of the present. Such a sentiment was expressed eloquently by the

German philosopher Günther Anders (in Schraube, 2005:78):

It is not enough to change the world. We do this anyway, and it mostly happens without our efforts, regardless. What we have to do is interpret these changes so we can in turn change the changes, so that the world doesn’t go on changing without us – and not ultimately become a world without us.

At this point, however, I want to briefly engage with some of the changes which have taken place outside the time and space analysed here and point to ways in which the thesis might speak to them.

The Rwandan genocide in 1994 sparked a destructive period of conflict at the heart of the continent that claimed millions of lives. An engagement with this continental warfare, which began in Congo in 1996 and lasted officially until 2002 – which Prunier (2009) has called ‘Africa’s World War’ – has proved beyond the remit of this thesis. Burundi and Rwanda were both heavily implicated in this conflict, and the Rwandan government, led by Paul

Kagame of the RPF, perpetrated massacres against Hutu refugees in camps in eastern Congo, discussed and investigated as acts of genocide (Newbury, 1998:8l; United Nations High

Commissioner for Human Rights, 2010; Lemarchand, 2011; Okosun and Kibiswa, 2013).

Consumers across the world were also implicated in this war, as it emerged that mines in the mineral-rich Kivu provinces of eastern Congo were being controlled by armed groups.

These groups operating in the de facto lawless space – which in Bauman’s (2002) terminology might be called a ‘global frontier-land’ – often under the auspices of neighbouring states like

96 See page 87. 261

Burundi, Rwanda and Uganda97, forged ‘territorial spheres of interest within which they could plunder Congo’s riches’ (Nzongola-Ntalaja, 2002:227). They also developed temporary alliances with each other (Reyntjens, 1999:246), often based on access to these territories.

The motivations for the parties entering war in 1996 were not economic. The Congo basin was a cauldron into which local and regional disputes spilled, such as those of Burundi and Rwanda. The control of mines, forests, borders and trade routes, however, soon became vital means of financing armed groups’ involvement (Nest, 2011:71). Minerals from this area, such as coltan, tin and tungsten, found their way into factories where products of the digital age were assembled before being dispatched to hi-tech areas of the world. Again, mass violence in the Great Lakes region of Africa is not a phenomenon apart from a globalised world.

In Burundi and Rwanda there have also been markedly different approaches to post- genocide reconstruction. Following the formal establishment of peace in 2006, Burundi embarked precariously on a programme of democratisation and ethnic power-sharing

(Vandenginste, 2014). In Rwanda, Paul Kagame’s government has been both praised for its remarkable development following the genocide in 1994 and reprimanded for its authoritarianism (Reyntjens, 2016). In Rwanda, as the Kigali genocide museum details, ethnic identities are deemed to have been European inventions that disrupted the harmonious relations of precolonial Rwandans. Recourse to these identities is a punishable offence. Victoire Ingabire

Umahoza, the chairperson of the Unified Democratic Forces (UDF), a coalition of opposition forces, has at the time of writing been under arrest since 2010 for suggesting that the non-Tutsi

97 The benefits to the states of Rwanda and Uganda are demonstrated in vast irregularities in export funds: Uganda, for example, exported $60 million in gold in 1996, and $105 million the next year. In 1995, it had exported a comparatively meagre $23 million (Prunier, 2009:156). Accordingly, Gérard Prunier said that the war was ‘the first known instance of postcolonial imperial conquest in Africa by an African country’ (2009:333). Today, Prunier has suggested, President Kagame of Rwanda ‘does not try to control the Congo anymore but simply to control enough mining interests in the Congo to help finance his great dreams of turning Rwanda into the Singapore of Africa’ (2009:326). As much is demonstrated by his direct support of the ‘M23’ rebellion, exposed by a recent UN Security Council letter (2012). 262

victims of the 1994 genocide be commemorated. While a truth and reconciliation committee in Burundi was being mooted (Nimuraba, 2014), attempts to bring perpetrators to justice have met with considerable difficulty, contrasting with the post-genocide legal proceedings of the

International Criminal Tribunal for Rwanda (ICTR) and the gacaca courts98.

Burundi became re-embroiled in a civil conflict in 2015 following the violation of the peace agreement by President Pierre Nkurunziza, who sought an unconstitutional third term in office. What began initially as protests in the name of democracy have given rise to grievances and ethnicised rhetoric harking back to previous rounds of conflict, raising fears about a return to mass conflict and even genocide (Ahluwalia, 2016; Daley and Popplewell, 2016). Since the conflict began, hundreds of thousands of refugees have fled the country into camps of neighbouring territories in Tanzania, Congo and Rwanda (Marima, 2015). Political opposition and civil society activists have been rounded up and detained in secretive prisons, torture is rife, and non-state media outlets have been shut down (International Crisis Group, 2016).

Further instability has resulted from events in neighbouring DR Congo, as president

Joseph Kabila strives to remain in power for a third term after delaying scheduled 2016 elections, which would violate the country’s constitution in a similar sense to Pierre

Nkurunziza in Burundi (Mclean and Burke, 2016). The eastern provinces have remained in a state of low-level conflict since the wars and have provided shelter for armed groups with grievances in Burundi, Rwanda and Uganda. It is uncertain what might happen in Rwanda if regional turmoil ensues from the problems of DR Congo and Burundi. There are rumours that

Rwanda is attempting to destabilise the Burundian government of Pierre Nkurunziza and is providing military training to Burundian refugees (Al Jazeera, 2016).

98 Gacaca translates as ‘on the grass’ in Kinyarwanda and refers to the centrepiece of Rwanda’s justice and reconciliation programme. It is a traditional community justice mechanism whereby trials take place in the open within communities. They have been received ambivalently (for discussion, see Amnesty International, 2002; Clark, 2010). 263

These developments should be of great concern. The analysis presented in this thesis is important for interpretation of the contemporary scene for a number of reasons. In the first instance, it demonstrates the necessity of understanding the divergent nature of the contemporary crises, especially of Burundi from Rwanda. In much contemporary reportage of

Burundi (Buchanan, 2016), the reference point for the potential for genocide is very often the experience of Rwanda, rather than Burundi’s own historical experiences of genocidal violence, which as we have seen are intimately tied up with conflicts that are specific to Burundian history. At the same time, the thesis has also demonstrated the ‘manifold interlockings’ (D.

Newbury, 1998:75) of post-independence violence in the Great Lakes region, nowhere more intimately entangled than in Burundi and Rwanda. There is great potential for mass conflict in the worsening regional situation of the Great Lakes. As detailed in this thesis, toxic mixtures of history and myth flourish in situations of fear, insecurity, impunity and crisis. It is in these situations, where previously unrecognised historical grievances come to the surface, that ethnicity can be mobilised for the purposes of violence. As such, it needs to be remembered that genocide is not the result of primordial ethnic hatreds.

The conclusions here about the relationship between modernity and violence, and about the ways in which traditional heritages can be allied to repressive forms of political rule, do not pertain solely to this immediate context. The various authoritarianisms of Nkurunziza, Kagame and Kabila, though specific, are not anomalous and cannot be reduced simply to a tendency towards corruption and anti-democratic illiberalism supposedly endemic to African politics.

They are perhaps emblematic of a more widespread emergence of authoritarian ‘strong men’ across the world, in countries as diverse as India, Turkey and Russia.

Over the period in which this thesis was completed, there has been in Western societies a sharp rise in xenophobia on the streets as well as in political discourse. This can be seen in the USA surrounding the rise and presidential election of Donald Trump, and in numerous 264

countries across Europe, most notably encapsulated in the ‘Brexit’ vote of June 2016 in the

UK. Accusations of ‘treachery’ are aimed at those who dissent or who insist on democratic and parliamentary scrutiny, who have been openly vilified in mainstream media as ‘enemies of the people’. These are traits of the ethnonationalist characteristics of the ‘dark side of democracy’

(Mann, 2004).

All of this is taking place amid the worst refugee crisis since World War Two and devastating conflict and instability in the Middle East following disastrous Western foreign policy interventions, as well as the re-emergence of Cold War hostilities and rivalries between power blocs. Underpinning this is what many have argued to be a terminal crisis of global neoliberal capitalism, and unfolding in the background is a climatological crisis attributable to irreversible global warming. If we have since the financial crash of 2007-8 been living, to quote

Antonio Gramsci (1971:556), in a crisis consisting ‘precisely in the fact that the old is dying and the new cannot be born’, it is prudent to wonder whether or not the ‘new’ is beginning to solidify. Following Trump’s election, an aide to Marine Le Pen, leader of the far-right Front

National party in France, suggested that ‘their world is collapsing, ours is being built’ (Irish

Times, 2016).

I do not wish make overwrought and untenable parallels between the specific histories of Burundi and Rwanda and these trends and problems. I suggest instead that the most pressing moral conclusion of this thesis is that ethno-nationalist violence of the sort seen to an extreme degree in Burundi and Rwanda cannot simply be ‘distanced’ as a peculiar problem of purportedly backwards and spatially isolated national societies. Rather, it ought to be seen as a possibility entangled in modernity and in principle extendable outside the specific context analysed here. Any complacency regarding the inevitable progress and peaceability of modernity ought to be checked in the contexts of the multiple crises of the contemporary world. 265

This is not equivalent to a nihilistic rejection of the principle that human action can project articulations of a better world from the ‘space of experience’ onto the ‘horizon of expectation’ (Koselleck, 2004 [1979]:255-276). Indeed it is this principle that enables us to speak of ‘crimes against humanity’, of genocide prevention, and of anti-colonialism. To paraphrase Michel Foucault (1982:231-232), we should say that it is not the case that modernity itself is bad, but that it contains possibilities that are dangerous. This position ‘leads not to apathy but to a hyper- and pessimistic activism… the ethico-political choice we have to make every day is to determine which is the main danger’. But by unduly distancing the experiences of Burundi and Rwanda, comforting ourselves that the sorts of violent social conflicts that we have seen there are far away and long ago, we achieve only an avoidance of confronting these dangers.

It has been said many times that, because it probes beneath our ‘common sense’ understandings of social reality, sociology points to possibilities that are latent in the present

(Bauman and May, 2001). These possibilities cannot be seen in purely positive terms. What a consideration of Burundi and Rwanda has shown here is that, though contingent on highly specific factors and experiences, trajectories to and through modernity, which are not separate and hermetic but entangled with the experience of Western modernity, can be manifested in violent social conflict. Violent social conflict is a possibility of the multiplicity of modernity and therefore it cannot be distanced as its counter-image.

This was the powerful message of Bauman’s Modernity and the Holocaust (1989). He suggested before writing that book that he had seen the Holocaust ‘like a picture on the wall: neatly framed, to set the painting apart from the wall paper and emphasise how different it was from the rest of the furnishings’. Instead, he realised, ‘the Holocaust was a window, rather than a picture on the wall’, and this window revealed fundamental dangers, the knowledge of which were ‘of the utmost importance not just for the perpetrators, victims and witnesses of the crimes, 266

but for all those who are alive today and hope to be alive tomorrow’ (Bauman, 1989:vii-viii).

The experiences of Burundi and Rwanda, similarly, ought not to be framed as a picture on the wall in the museum of ‘man’s inhumanity to man’99. Instead, cases like Burundi and Rwanda also ought to be seen as a window into some dark potentialities of modernity.

Following the late Ulrich Beck, Bauman (2016) has spoken of how the reality of cosmopolitan conditions, wrought by de facto planetary interdependence, has yet to be matched by cosmopolitan awareness. The feeling of ‘co-presence’ on a shared planet does not automatically entail any ethical commitment and acknowledgement of a shared ‘world’. On the contrary, there seems to have emerged a broad consensus that tackling global problems which affect humanity tout court can be ignored in favour of retreats into national isolation, based on exceptionalist, melancholic evocations of supposed golden eras. Walls and fences are being constructed on national boundaries, in the spaces of intercultural horizons, decades after the collapse of the Berlin Wall heralded a purported ‘end of history’ and the supposed inauguration of an era of unprecedented openness and mobility. But, as Bauman (2008:75-77) urges, there can be no national solutions to global problems. Enormous population movements requiring some ethical and political responses are bound to ensue, and indeed, have already begun, not only from conflict and poverty and their conflation, but from transformations in regional ecologies. The poorest of the poor are disproportionately affected by such risks (Beck, 2009).

The paradigm of the multiplicity of modernity, particularly as represented in the exceptional historical and geographical scope of Eisenstadt’s work, demonstrates the importance of the discipline of sociology and its traditional moorings in the development of a normative, historically-sensitive cosmopolitan awareness. Such an awareness, as Delanty

(2006a:25) notes, is itself ‘integral to modernity, insofar as this is a condition of self-

99 Gadamer (2004 [1975]:135) shared with Bauman an appreciation of how framing events like a picture to be hung up on the wall detaches them from their ‘connections with life and the particular conditions of our approach to it’. 267

problematization, incompleteness and the awareness that certainty can never be established once and for all’; within such a condition, sociology possesses the specific ability to ‘discern or make sense of social transformation by identifying new or emergent social realities’ (Delanty,

2006a:25). If modernity unfolds non-teleologically along specific, entangled trajectories or routes, this also means that it must necessarily be put ‘on endless trial’, to use Leszek

Kołakowski’s (1990) phrase. In the same sense that, as Gadamer (2004 [1975]:302) put it, ‘to be historically means that knowledge of oneself can never be complete’, to understand modernity historically means that knowledge of it can never be complete and that it should be subject to continuous, deep and reflexive scrutiny. This requires a patient understanding of broader patterns of social change and the possibilities immanent in them.

268

BIBLIOGRAPHY

Abrams, Jason. 1995. Burundi: Anatomy of an Ethnic Conflict. Survival: Global Politics and Strategy, 37 (1), 144-164. Abrams, Philip. 1980. History, Sociology, Historical Sociology. Past and Present, 87, 3-16. Abrams, Philip. 1982. Historical Sociology. Shepton Mallet: Open Books. Achebe, Chinua. 1988. An Image of Africa: Racism in Conrad’s Heart of Darkness. In Kimbrough, Robert (Ed.) Heart of Darkness: An Authoritative Text, Background and Sources, Criticism. London: W.W. Norton and Co., 251-261 Adams, Suzi. 2009. The Intercultural Horizons of Johann P. Arnason’s Phenomenology of the World. Journal of Intercultural Studies, 30 (3), 249-266. Adorno, T. W. 2001. The Culture Industry: Selected Essays on Mass Culture. London: Routledge. Adorno, T. W. & Horkheimer, Max. 1997 [1944]. Dialectic of Enlightenment. London: Verso. African Rights. 1994. Rwanda: Death, Despair, and Defiance. London: African Rights. Agamben, Giorgio. 1998. Homo Sacer: Sovereign Power and Bare Life. Stanford, CA: Stanford University Press. Ahluwalia, Pal. 2016. Burundi: We Cannot Allow Another Genocide. African Identities, 14 (1), 1-2. Albert, Ethel. 1960. Socio-Political Organization and Receptivity to Change: Some Differences between Ruanda and Urundi. Southwestern Journal of Anthropology, 16 (1), 46-74. Alexander, Jeffrey. 2006. The Civil Sphere. Oxford: Oxford University Press. Alexander, Jeffrey. 2009. The Social Construction of Moral Universals. In Alexander (ed.) Remembering the Holocaust: a Debate. Oxford: Oxford University Press, 3-104. Alexander, Jeffrey. 2012. Trauma: A Social Theory. Cambridge: Polity. Alexander, Jeffrey. 2013. The Dark Side of Modernity. Cambridge: Polity. Alexander, Jeffrey. 2016. Cultural Trauma, Morality and Solidarity: The Social Construction of ‘Holocaust’ and Other Mass Murders. Thesis Eleven, 132 (1), 3-16. Al Jazeera. 2016. Rwandan FM: Rwanda is not ‘destabilising’ Burundi: Louise Mushikiwabo denies allegations that the government provided military training to Burundian refugees. Al Jazeera. Available online at http://www.aljazeera.com/programmes/upfront/2016/06/rwandan-fm- rwanda-destabilising-burundi-160611070014188.html Amelina, Anna (ed.). 2012. Beyond Methodological Nationalism: Research Methodologies for Cross-Border Studies. London: Routledge. Amnesty International. 1992, May. Rwanda: Persecution of Tutsi Minority and Repression of Government Critics, 1990-1992. Anderson, Regine. 2000. How Multilateral Development Assistance Triggered the Conflict in Rwanda. Third World Quarterly, 21 (3), 441-456. Anstey, Roger. 1966. King Leopold’s Legacy: The Congo under Belgian Rule, 1908-1960. London: Oxford University Press for the Institute of Race Relations. Arendt, Hannah. 1967 [1951]. The Origins of Totalitarianism. London: George Allen & Unwin Ltd. Arendt, Hannah. 1990 [1963]. On Revolution. London: Penguin. Arendt, Hannah. 1994. Essays in Understanding, 1930-1954: Formation, Exile and Totalitarianism. New York, NY: Schocken Books. Arendt, Hannah. 1998 [1958]. The Human Condition. Chicago, IL: University of Chicago Press. Arjomand, Saïd Amir. 2014. Introduction: The Challenge of Integrating Social Theory and Regional Studies. In Arjomand (ed.) Social Theory and Regional Studies in the Global Age. Albany, NY: SUNY Press, 1-22. Arnason, J. P. 1991. Modernity as Project and as Field of Tensions. In Honneth & Joas (eds.) Communicative Action: Essays on Jürgen Habermas’s ‘The Theory of Communicative Action’. Cambridge, MA: MIT Press, 181-213. 269

Arnason, J. P. 1993. The Future that Failed: Origins and Destinies of the Soviet Model. London: Routledge. Arnason, J. P. 1996. Invention and Emergence: Reflections of Hans Joas Theory of Creative Action. Thesis Eleven, 47, 101-113, Arnason, J. P. 2000. Communism and Modernity. Daedalus, 129 (1), 61-90. Arnason, J. P. 2001. The Multiplication of Modernity. In Ben-Rafael & Sternberg (eds.) Identity, Culture and Globalisation. Leiden: Brill, 131-154. Arnason, J. P. 2003. Civilizations in Dispute: Historical Questions and Theoretical Traditions. Leiden: Brill. Arnason, J. P. 2003a. Entangled Communisms: Imperial Revolutions in Russia and China. European Journal of Social Theory, 6 (3). 307-325. Arnason, J. P. 2003b. Sociology, Critique and Modernity: Views across the European Divide. In Ben-Rafael (ed.) Sociology and Ideology. Leiden: Brill. Arnason, J. P. 2005. Alternating Modernities: The Case of Czechoslovakia. European Journal of Social Theory, 8 (4), 435-451. Arnason, J. P; Blokker, Paul & Delanty, Gerard. 2011. An Interview with Johann P. Arnason: Critical Theory, Modernity, Civilizations and Democracy. European Journal of Social Theory, 14 (1), 119-132. Arnason, J. P., Eisenstadt, S. N. & Wittrock, Bjorn (eds.). 2005. Axial Civilisations and World History. Leiden: Brill. Ascione, Gennaro & Chambers, Ian. 2016. Global Historical Sociology: Theoretical and Methodological Issues – an Introduction. Cultural Sociology [online first version], 1-16. Available online at: http://cus.sagepub.com/content/early/2016/05/04/1749975516639086.full.pdf+html Badiou, Alain. 2002. Ethics: An Essay on the Understanding of Evil, trans. Hallward. London: Verson. Baker, J. N. L. 1962. John Hanning Speke. The Geographical Journal, 78 (4), 385-388. Baker, Randall. 1970. Reorientation in Rwanda. African Affairs, 69 (275), 141-154. Balandier, Georges. 1966 [1951]. The Colonial Situation: A Theoretical Approach. In Wallerstein (ed.) Social Change: The Colonial Situation. New York, NY: John Wiley & Sons Inc., 34-62. Balandier, Georges. 1968. Daily Life in the Kingdom of the Kongo: From the Sixteenth to the Eighteenth Century. London: George Allen & Unwin Ltd. Balandier, Georges. 1970. The Sociology of Black Africa: Social Dynamics in Central Africa, trans. Douglas Garman. London: Andre Deutsche. Ballantyne, Glenda. 2008. Unification or Multiplication? Theorizing Multiple Modernities. The International Journal of the Humanities, 5 (10), 53-59. Barns, T. A. 1923. The Wonderland of Eastern Congo: The Region of the Snow-Crowned Volcanoes, the Pygmies, the Giant Gorilla, and the Okapi. London: G. P. Putnam’s Sons. Basaninyenzi, Gatsinzi. 2006. ‘Dark-Faced Europeans’: The Nineteenth-Century Colonial Travelogue and the Invention of the Hima Race. In Young & Braziel (eds.) Race and the Foundations of Knowledge: Cultural Amnesia in the Academcy. Chicago IL: University of Chicago Press, 114-126. Bauer, Yehuda. 2001. Rethinking the Holocaust. New Haven, CT: Yale University Press. Bauman, Zygmunt. 1978. Hermeneutics and Social Science. London: Hutchinson & Co. Bauman, Zygmunt. 1985. Stalin and the Peasant Revolution: A Case Study in the Dialectics of Master and Slave, Leeds Occasional Papers in Sociology, 19. Bauman, Zygmunt. 1987. Legislators and Interpreters: On Modernity, Postmodernity and Intellectuals. Cambridge: Polity. Bauman, Zygmunt. 1989. Modernity and the Holocaust. Cambridge: Polity. 270

Bauman, Zygmunt. 1990. Modernity and Ambivalence. In Featherstone (ed.) Global Cultureː Nationalism, Globalization and Modernity. ʟondonː SAGE, 143-164. Bauman, Zygmunt. 1991. Modernity and Ambivalence. Cambridge: Polity. Bauman, Zygmunt. 1995. Life in Fragments: Essays in Postmodern Morality. Oxford: Blackwell. Bauman, Zygmunt. 1997. Modernity and Clarityː the Story of a Failed Romance. In Luthe & Wiedenmann (eds.) Ambivalenzː Studien zum kulturtheoretischen und empirischen Gehalt einer Kategorie der Erschließung des Unbestimmten. VS Verlag für Sozialwissenschaften, 109-122. Bauman, Zygmunt. 1998. Globalization: The Human Consequences. Cambridge: Polity. Bauman, Zygmunt. 2000. Liquid Modernity. Cambridge: Polity. Bauman, Zygmunt. 2002. Reconnaissance Wars of the Plantetary Frontierland. Theory, Culture & Society, 19 (4), 81-90. Bauman, Zygmunt. 2004. Stalin. Cultural Studies, 4 (1), 3-11. Bauman, Zygmunt. 2008. Does Ethics Have a Chance in a World of Consumers? Cambridge, MA: Harvard University Press. Bauman, Zygmunt. 2010. Living on Borrowed Time: Conversations with Citlali Rovirosa-Madrazo. Cambridge: Polity. Bauman, Zygmunt. 2016, July 27. No More Walls in Europe: Tear Them Down! Social Europe. Available online at https://www.socialeurope.eu/2016/07/no-walls-europe-tear/ Bauman, Zygmunt & May, Tim. 2001. Thinking Sociologically, 2nd ed. Oxford: Blackwell. Bauman, Zygmunt & Weltzer, Harald. 2002. On the Rationality of Evil: An Interview with Zygmunt Bauman. Thesis Eleven, 70, 100-112. Bayart, J. F. 2009 [1986]. The State in Africa: The Politics of the Belly, 2nd ed. Cambridge: Cambridge University Press. BBC News. 1994, April 6th. Rwanda presidents’ plane ‘shot down’. Available online at: http://news.bbc.co.uk/onthisday/hi/dates/stories/april/6/newsid_2472000/2472195.stm BCRUIPRO (Belgian Congo and Ruanda-Urundi Information and Public Relations Office). 1960. Ruanda-Urundi: Geography and History. Brussels: Belgian Congo and Ruanda-Urundi Information and Public Relations Office. Beck, Ulrich. 2007. The Cosmopolitan Condition: Why Methodological Nationalism Fails. Theory, Culture & Society, 24 (7-8), 286-290. Beck, Ulrich. 2009. World at Risk. Cambridge: Polity. Beck, Ulrich & Beck-Gernsheim, Elisabeth. 2002. Individualism: Institutionalised Individualism and its Social and Political Consequences. London: SAGE. Beilharz, Peter. 2002. Modernity and Communism: Zygmunt Bauman and the Other Totalitarianism. Thesis Eleven, 70 (1), 88-99. Bellah, Robert & Joas, Hans (eds.). 2012. The Axial Age and its Consequences. Cambridge, MA: Harvard University Press. Berardi, Franco. 2011. After the Future. Oakland, CA: AK Press. Berman, Bruce. 1984. Structure and Process in the Bureaucratic States of Colonial Africa. Development and Change, 15 (2), 161-202. Berman, Bruce. 1997. The Perils of Bula Matari: Constraint and Power in the Colonial State. Canadian Journal of African Studies, 31 (3), 555-570. Bernstein, Richard. 1983. Beyond Objectivism and Relativism: Science, Hermeneutics, and Praxis. Oxford: Blackwell. Bernstein, Richard. 1986. The Rage against Reason. Philosophy and Literature, 10 (2), 186-210. Bernstein, Richard. 2002. Radical Evil: A Philosophical Interrogation. Cambridge: Polity. Bernstein, Richard. 2013. Violence: Thinking Without Bannisters. Cambridge: Polity. Bhambra, Gurminder. 2007. Rethinking Modernity: Postcolonialism and the Sociological Imagination. Basingstoke: Palgrave. 271

Bhambra, Gurminder. 2007a. Sociology and Postcolonialism: Another ‘Missing’ Revolution? Sociology, 41 (5), 871-884. Bhambra, Gurminder. 2011. Talking among Themselves? Weberian and Marxist Historical Sociologies as Dialogues without ‘Others’. Millennium Journal of International Studies, 39 (3), 667-681. Bhambra, Gurminder. 2014. Connected Sociologies. London: Bloomsbury. Bhambra, Gurminder. 2016. Comparative Historical Sociology and the State: Problems of Method. Cultural Sociology, 10 (3), 335-351. Bindseil, Reinhart. 1992. Ruanda im Lebensbild des Offiziers, Afrikafoschers und Kaiserlichen Gouverneurs Gustav Adolf Graf von Götzen, 1866-1910. Berlin: Dietrich Reimer Verlag. Bishikwabo, Chubaka. 1991. Institutions coutumières et modernisme au Zaïre. In Département d’histoire de l’univesité du Burundi (eds.) Histoire sociale de l’Afrique de l’Est (XIXe-XXe siècle): Actes du Colloque de Bujumbura (17-24 octobre 1989). Paris: Karthala, 15-35. Bleicher, Josef. 1980. Contemporary Hermeneutics: Hermeneutics as Method, Philosophy and Critique. London: Routledge. Bloxham, Donald. 2007. The Great Game of Genocide: Imperialism, Nationalism, and the Destruction of the Ottoman Armenians. Oxford: Oxford University Press. Bloxham, Donald. 2008. Modernity and Genocide. European History Quaterly, 38 (2), 294-311. Bloxham, Donald. 2009. The Final Solution: A Genocide. Oxford: Oxford University Press. Bloxham, Donald & Moses, A. D. (eds.). 2010. The Oxford Handbook of Genocide Studies. Oxford: Oxford University Press. Boatcă, Manuela. 2014. Second Slavery versus Second Serfdom: Local Labor Regimes of the Global Periphery. In Arjomand (ed.) Social Theory and Regional Studies in the Global Age. Albany, NY: SUNY Press, 361-388. Boltanski, Luc. 1999. Distant Suffering: Morality, Media and Politics. Cambridge: Cambridge University Press. Boltanski, Luc & Chiapello, Eve. 2007. The New Spirit of Capitalism. London: Verso. Botte, Roger. 1985. Rwanda and Burundi: Chronology of a Slow Assassination, part I. The International Journal of African Historical Studies, 18 (1), 53-91. Botte, Roger. 1985a. Rwanda and Burundi: Chronology of a Slow Assassination, part II. The International Journal of African Historical Studies, 18 (2), 289-314. Boyer, Allison. 1992. Burundi: Unity at Last? Africa Report, 37 (2), 37. Braudel, Fernand. 1995 [1949]. The Mediterranean and the Mediterranean World in the Age of Philip II, vol. II. Berkley, CA: University of California Press. Brecht, Arnold. 1959. Political Theory: The Foundations of Twentieth-Century Political Thought. Princeton, NJ: Princeton University Press. Breed, Ananda. 2006. Performing Reconciliation in Rwanda. Peace Review, 18 (4), 507-513. Brooke, James. 1987, June 4th. Rule by Minority Persists in Burundi. New York Times, 12. Bryant, Joseph. 2000. On Sources and Narratives in Historical Social Science: a Realist Critique of Postitivist and Postmodernist Epistemologies. British Journal of Sociology, 51 (3), 489-533. Buber, Martin. 1992. On Intersubjectivity and Cultural Creativity. Chicago, IL: University of Chicago Press. Buchanan, Elsa. 2016, March 29th. Burundi: ‘A genocide is being prepared and the UN will be too late’. IB Times. Available online at: http://www.ibtimes.co.uk/burundi-genocide-being- prepared-un-will-be-too-late-1552054 Bunge, Mario & Mahner, Mario. 1997. Foundations of Biophilosophy. Berlin: Springer-Verlag. Burawoy, Michael. 2008. For a Subaltern Global Sociology? Current Sociology, 56 (3), 435-444. Burke, Peter. 1993. History and Social Theory. Ithaca, NY: Cornell University Press. Carney, James. 2011. From Democratization to Ethnic Revolution: Catholic Politics in Rwanda, 1950-1962. Unpublished PhD Thesis: Catholic University of America. 272

Carruthers, Susan. 2004. Tribalism and tribulation: media constructions of ‘African savagery’ and ‘Western humanitarianism’ in the 1990s. In Allan, S & Zelizer, B (Eds.), Reporting War: Journalism in Wartime. London: Routledge, 155-174. Castoriadis, Cornelius. 1987 [1975]. The Imaginary Institution of Society, trans. Blamey. Cambridge: Polity Press. Castoriadis, Cornelius. 1991. Philosophy, Politics, Autonomy, ed. Curtis. Oxford: Oxford University Press. Castoriadis, Cornelius. 1997. World in Fragments: Writings on Politics, Society, Psychoanalysis and the Imagination. Stanford, CA: Stanford University Press. Caton-Jones, Michael (dir.). 2006. Shooting Dogs [film]. London: BBC Films. Césaire, Aimé. 2000 [1972]. Discourse on Colonialism. New York, NY: Monthly Review Press. Chakrabarty, Dipesh. 2009. Provincializing Europe: Postcolonial Thought and Historical Difference. Princeton, NJ: Princeton University Press Chalk, Frank & Jonassohn, Kurt. 1990. The History and Sociology of Genocide: Analyses and Case Studies. New Haven, CT: Yale University Press. Chatterjee, Partha. 1986. Nationalist Thought and the Colonial World: A Derivative Discourse. London: Zed Books. Chernilo, Daniel. 2011. The Critique of Methodological Nationalism: Theory and History. Thesis Eleven, 106 (1), 98-117. Chrétien, Jean-Pierre. 1968. Le passage de l’expédition d’Oscar Baumann au Burundi (Septembre- Octobre 1892). Cahiers d’Études Africaines, 8 (29), 48-95. Chrétien, Jean-Pierre. 1970. Un révolte au Burundi en 1934: Les raciness traditionalistes de l’hostilité a la colonisation. Annales. Histoires, Sciences Sociales, 25 (6), 1648-1717. Chrétien, Jean-Pierre. 1972. Le révolte de Ndungutse (1912). Forces traditionelles et pression colonial au Rwandan allemande. Revue française d’histoire d’outre-mer, 59 (4), 645-680. Chrétien, Jean-Pierre. 1990. Social Sciences Facing Ethnic Violence. Issue: A Journal of Opinion, 19 (1), 38-40. Chrétien, Jean-Pierre. 1997. Le défi de l’ethnisme: Rwanda et Burundi, 1990-1996. Paris: Karthala. Chrétien, Jean-Pierre. 2003. The Great Lakes of Africa: Two Thousand Years of History, trans. Straus. New York, NY: Zone Books. Chrétien, Jean-Pierre. 2010. L’invention de l’Afrique des Grands Lacs: Une histoire du XXe siècle. Paris: Karthala. Chrétien, Jean-Pierre & Bahenduzi, Michel. 1990. Ntare Rushatsi est-il passé à Magamba en mars 1680 ou en août 1701? Culture et société: Revue de civilisation burundaise, 11 (1), 38-55. Chrétien, Jean-Pierre & Dupaquier, Jean-François. 2007. Burundi 1972: Au bord des genocides. Paris: Karthala. Christensen, Hanne. 1985. Refugees and Pioneers: History and Field Study of a Burundian Settlement in Tanzania. Geneva: United Nations Research Institute for Social Development. Chossudovsky, Michel. 1996. Economic Genocide in Rwanda. Economic and Political Weekly, 31 (15), 938-941. Chouliaraki, Lilie. 2006. The Spectatorship of Suffering. London: Pine Forge Press. Clark, Timothy. 2002. Martin Heidegger. London: Routledge. Clark, Phil. 2010. The Gacaca Courts, Post-Genocide Justice and Reconciliation in Rwanda: Justice without Lawyers. Cambridge: Cambridge University Press. Classe, Léon. 1916. Notes rédigées par le R.P. Classe, des Pères Blancs, Mission de KABGAYI, à la demande de l’Administration Belge. 28 août, 11 pages. L’Organisation Politique du Rwanda au début de l’Occupation Belge: Jean-Marie Derscheid Collection Codere, Helen. 1962. Power in Ruanda. Anthropologica, 4 (1), 45-85. 273

Cohen, D. W. 1989. Peoples and States of the Great Lakes Region. In Ade Ajayi (ed.) General History of Africa, vol. 6: Africa in the Ninteenth Century until the 1880s. Paris: UNESCO, 270-293. Cohen, Robin & Kennedy, Paul. 2013. Global Sociology (3rd ed.). Basingstoke: Palgrave Macmillan. Coker, Christopher. 2015. Future War. Cambridge: Polity. Comaroff, John, 1998. Reflections on the Colonial State, in South Africa and Elsewhere: Factions, Fragments, Facts and Fictions. Social Identities: Journal for the Study of Race, Nation and Culture, 4 (3), 321-361. Comaroff, John. 2002. Governmentality, Materiality, Legality, Modernity: On the Colonial State in Africa. In Deutsche, Probst & Schmidt (eds.) African Modernities: Entangled Meanings in Current Debate. Oxford: James Currey Ltd, 107-134. Connell, Raewyn. 2007. Southern Theory: The Global Dynamics of Knowledge in Social Science. Sydney: Allen & Unwin. Conrad, Joseph. 1997 [1903]. Heart of Darkness. London: Penguin Popular Classics. Convention on the Prevention and Punishment of the Crime of Genocide (General Assembly Resolution 260 III). 9th December, 1948. Available online at http://www.un.org/ga/search/view_doc.asp?symbol=a/res/260(III) Cordero, Rodrigo. 2017. Crisis and Critique: On the Fragile Foundations of Social Life. London: Routledge. Costa, Sérgio. 2007. ‘Deprovincializing’ Sociology: The Postcolonial Contribution. Revista Brasileira de Ciencias Sociais, 3, 1-20. Daley, Patricia. 2006. Ethnicity and Political Violence in Africa: The Challenge to the Burundi State. Political Geography, 25 (1), 657-679. Daley, Patricia. 2008. Gender and Genocide in Burundi: The Search for Spaces of Peace in the Great Lakes Region. Oxford: James Currey. Daley, Patricia & Popplewell, Rowan. 2016. The Appeal of Third Termism and Militarism in Burundi [electronic version]. Review of African Political Economy. Available online at http://dx.doi.org/10.1080/03056244.2015.1111202 Daskalov, Roumen. 2013. Preface. In Daskalov & Marinov (eds.) Entangled Histories of the Balkans: Vol I. National Ideologies and Language Policies. Leiden: Brill, xvii – xv. Dawson, Matt. 2016. Social Theory and Alternative Futures. Basingstoke: Palgrave Macmillan. De Lame, Danielle. 2004. Mighty Secrets, Public Commensality, and the Crisis of Transparency: Rwanda through the Looking Glass. Canadian Journal of African Studies, 38 (2), 279-317. Delanty, Gerard. 2006. Modernity and the Escape from Eurocentrism. In Delanty (ed.) Handbook of Contemporary European Social Theory, London: Routledge, 266-278. Delanty, Gerard. 2006a. The Cosmopolitan Imagination: Critical Cosmopolitanism and Social Theory. British Journal of Sociology, 57 (1), 25-47. Delanty, Gerard. 2008. Civilisational Constellations and Modernity. In Delanty (ed.) Europe and Asia beyond East and West. London: Routledge, 45-60. Delanty, Gerard. 2013. Formations of European Modernity: A Historical and Political Sociology of Europe. Basingstoke: Palgrave Macmillan. Delanty, Gerard & Isin, Engin. 2003. Reorienting Historical Sociology. In Delanty & Isin (eds.) Handbook of Historical Sociology. London: SAGE, 1-8. Delanty, Gerard & Mota, Aurea. 2015. Eisenstadt, Brazil and the Multiple Modernities Framework: Revisions and Reconsiderations. Journal of Classical Sociology, 15 (1), 39-57. De Juan, Alexander. 2015. State Extraction and Anti-Colonial Rebellion: Quantitative Evidence from the Former German East Africa. German Institution of Global and Area Studies, Working Paper no. 271. Available online at: www.giga‐hamburg.de/workingpapers De Meyer, Roger. 1958. Introducing the Belgian Congo and the Ruanda-Urundi. Brussels: Office de Publicité, S. A. 274

Deringil, Selim. 2003. ‘They Live in a State of Nomadism and Savagery’: The Late Ottoman Empire and the Post-Colonial Debate. Comparative Studies in Society and History, 45 (2), 311-342. De Swaan, Abram. 2001. Dyscivilization, Mass Extermination and the State. Theory, Culture & Society. 18 (2), 265-276. De Swaan, Abram. 2015. The Killing Compartments: The Mentality of Mass Murder. New Haven, CT: Yale University Press. De Vleeschauwer, Albert. 1943. Belgian Colonial Policy. New York, NY: Belgian Information Centre. De Waal, Alex & Conley-Zilkic, Bridget. 2006. Reflections on How Genocidal Killings are Brought to an End. Social Science Research Council: How Genocides End. Available online at: http://howgenocidesend.ssrc.org/ Des Forges, Alison. 1995. The Ideology of Genocide. Issue: A Journal of Opinion, 23 (2), 44-47. Des Forges, Alison. 1999. Leave None to Tell the Story: Genocide in Rwanda. New York, NY: Human Rights Watch. Des Forges, Alison. 2011. Defeat is the Only Bad News: Rwanda under Musinga, 1896-1931. Madison, WI: University of Wisconsin Press. Dewey, John. 1929. The Quest for Certainty: A Study of the Relation of Knowledge and Action. New York, NY: Minton, Balch & Company. Diamond, Jared. 2010. Collapse: How Societies Choose to Fail or Succeed. London: Penguin. Diggs. Rebecca. 2010. Under-Standing: The Divine Messenger Hermes and the Character of Hermeneutics. Mythological Studies Journal, 1 (1), 1-11. Diop, Cheikh Anta. 1989 [1955]. The African Origin of Civilization: Myth or Reality, trans. Cook. Chicago, IL: Lawrence Hill Books. Dirlik, Arif. 2003. Global Modernity? Modernity in an Age of Global Capitalism. European Journal of Social Theory, 6, 275-292. Domingues, José Maurício. 2009. Global Modernisation, ‘Coloniality’ and a Critical Sociology for Contemporary Latin America. Theory, Culture & Society, 26 (1), 112-133. Dowden, Richard. 1994, 10th April. Rwanda’s twins locked in eternal war; the Hutu and Tutsi are old foes bound together by blood – and by colonial borders that are tearing a region apart. The Independent, 12. Dowden, Richard. 2008. Africa: Altered States, Ordinary Miracles. London: Portobello Books. Driver, Felix. 1991. Henry Morton Stanley and His Critics: Geography, Exploration and Empire. Past & Present, 133 (1), 134-166. Duarte, Mary. 1995. Education in Ruanda-Urundi, 1946-61. Historian, 57 (2), 275-284. Du Bois, W. E. B. 1965 [1947]. The World and Africa: An Inquiry into the Part Which Africa has played in World History. International Publishers Co. Dunn, James. 2013. Genocide in East Timor. In Totten & Parsons (eds.) Centuries of Genocide: Essays and Eyewitness Accounts, 4th ed. London: Routledge, 279-317. Dunning, Eric & Mennell, Stephen. 1998. Elias on Germany, Nazism and the Holocaust: On the Balance between ‘Civilizing’ and ‘Decivilizing’ Trends in the Social Development of Western Europe. The British Journal of Sociology, 49 (3), 339-357. Durkheim, Émile. 2001 [1912]. The Elementary Forms of Religious Life, trans. Cosman. Oxford: Oxford University Press. Durkheim, Émile. 2006 [1897] On Suicide, trans. Buss. London: Penguin. Dussel, Enrique. 1993. Eurocentrism and Modernity. Boundary, 20 (3), 65-76. Dussel, Enrique. 1996. The Underside of Modernity: Apel, Ricoeur, Rorty, Taylor, and the Philosophy of Liberation, trans. Mendieta (Atlantic Highlands, NJ: Humanities Press. Eggers, Ellen. 2006. The Historical Dictionary of Burundi, 3rd ed. Lanham, MD: Scarecrow Press. Eisenstadt, S. N. 1949. The Relations between Sociological Theory and Anthropological Research. Man, 49 (1), 121-124. 275

Eisenstadt, S. N. 1954. Afican Age Groups: A Comparative Study. Africa, 24 (2), 100-113. Eisenstadt, S. N. 1965. Social Change and Modernization in African Societies South of the Sahara. Cahiers d’Études Africaines, 5 (19), 453-471. Eisenstadt, S. N. 1969. Some Observations on the Dynamics of Traditions. Comparative Studies in Society and History, 11 (4), 451-475. Eisenstadt, S. N. 1972. Intellectuals and Tradition. Daedalus, 101 (2), 1-19. Eisenstadt, S. N. 1973. Tradition, Change and Modernity. New York, NY: John Wiley & Sons. Eisenstadt, S. N. 1973a. Post-Traditional Societies and the Continuity and Reconstruction of Tradition. Daedalus, 102 (1), 1-27. Eisenstadt, S. N. 1996. Japanese Civilisation: A Comparative View. Chicago, IL: University of Chicago Press. Eisenstadt, S. N. 1996a. Barbarism and Modernity. Society, 33 (4), 31-39. Eisensatdt, S. N. 1998. Modernity and the Construction of Collective Identity. International Journal of Comparative Sociology. 39 (3), 138-158. Eisenstadt, S. N. 1999. Fundamentalism, Sectarianism, and Revolution: The Jacobin Dimension of Modernity. Cambridge: Cambridge University Press. Eisenstadt, S. N. 1999a. Multiple Modernities in an Age of Globalisation. Canadian Journal of Sociology, 24 (2), 283-295. Eisenstadt, S. N. 2000. Multiple Modernities. Daedalus, 129 (1), 1-29. Eisenstadt, S. N. 2001. The Civilisational Dimension of Modernity: Modernity as a Distinct Civilisation. International Sociology, 16 (3), 320-340. Eisenstadt, S. N. 2003. Comparative Civilizations and Multiple Modernities. Leiden: Brill. Eisenstadt, S. N. 2004. Pluralism and the Multiple Forms of Modernity (Interviewed by Delanty). European Journal of Social Theory, 7 (3), 391-404. Eisenstadt, S. N. 2005. Modernity in Socio-Historical Perspective. In Ben-Rafael & Sternberg (eds.) Comparing Modernities: Pluralism Versus Homogeneity: Essays in Homage to Schmuel N. Eisenstadt. Leiden: Brill, 31-56. Eisenstadt, S. N. 2005a. Collective Identity and the Constructive and Destructive Forces of Modernity. In Ben-Rafael & Sternberg (eds.) Comparing Modernities: Pluralism Versus Homogeneity: Essays in Homage to Schmuel N. Eisenstadt. Leiden: Brill, 635-657. Eisenstadt, S. N. & Lemarchand, René. 1981. Political Clientelism, Patronage and Development. London: SAGE. Eliade, Mircea. 1959. Cosmos and History: The Myth of the Eternal Return. New York, NY: Harper Torchbooks. Elias, Norbert. 1978. What is Sociology? trans. Mennel & Morrisey. London: Hutchinson & Co. Elias, Norbert. 1987. The Retreat of Sociologists into the Present. Theory, Culture & Society, 4 (1), 223-247. Elias, Norbert. 1994 [1939]. The Civilizing Process: The History of Manners and State Formation and Civilization. Oxford: Blackwell. Elias, Norbert. 2010 [1985]. The Loneliness of the Dying; and Humana Conditio: Observations on the Development of Humanity on the Fortieth Anniversary of the End of a War (8 May 1985), trans. Jephcott. In Scott & Scott (eds.) The Collected Works of Norbert Elias, vol. 6. Dublin: University College Dublin Press. Elias, Norbert. 2013 [1989]. Studies on the Germans, trans. Mennell & Dunning. In Mennell & Dunning (eds.) The Collected Works of Norbert Elias, vol. 11. Dublin: Dublin University Press. Elliot, G. F. S. 1896. A Naturalist in Mid-Africa: Being an account of a journey to the mountains of the moon and Tanganyika. London: A. D. Innes & Co. Elliott, Jane. 2005. Using Narrative in Social Research: Qualitative and Quantitative Approaches. London: SAGE. 276

Ellul, Jacques. 1978. The Betrayal of the West, trans. Matthew J. O’Connell. New York, NY: Continuum. Eltringham, Nigel. 2006. ‘Invaders who have stolen the country’: The Hamitic Hypothesis, Race and the Rwandan Genocide. Social Identities: Journal for the Study of Race, Nation and Culture, 12 (4), 425-446. Escobar, Araturo. 2007. Worlds and Knowledges Otherwise: The Latin American Modernity/Coloniality Research Programme. Cultural Studies, 21(2), 179-210. Evans, Brad & Giroux, Henry. 2015. Disposable Futures: The Seduction of Violence in the Age of Spectacle. San Francisco, CA: City Lights Books. Ewans, Martin. 2003. Belgium and the Colonial Experience. Journal of Contemporary European Studies, 11 (2), 167-180. Fabian, Johannes. 1983. Time and the Other: How Anthropology Makes its Object. New York, NY: Columbia University Press. Fabian, Johannes. 2000. Out of Our Minds: Reason and Madness in the Exploration of Central Africa. Berkeley, CA: University of California Press. Fackenheim, Emil. 1978. The Jewish Return into History: Reflections on the Age of Auschwitz and a New Jerusalem. New York, NY: Schocken. Fanon, Frantz. 1965. The Wretched of the Earth, trans. Constance Farrington. London: Macgibbon & Kee. Fanon, Frantz. 1967 [1952]. Black Skin, White Masks. Pluto Press: London. Feierstein, Daniel. 2011. Leaving the Parental Home: An Overview of the Current State of Genocide Studies. Genocide Studies and Prevention, 6(3), pp.257-269. Feierstein, Daniel. 2014. Genocide as Social Practice: Reorganising Society under the Nazis and Argentina’s Military Juntas. New Brunswick, NJ: Rutgers University Press. Fein, Helen. 1993. Genocide: A Sociological Perspective. London: SAGE. Fieldhouse, D. K. 1983. Colonialism, 1870-1945: An Introduction. Basingstoke: Macmillan. Fieldhouse, D. K. 1991. French and British Colonialism. Canadian Journal of African Studies / Revue canadienne des études africaines, 25 (1), 117-121. Finkelstein, Norman. 2000. The Holocaust Industry: Reflections on the Exploitation of Jewish Suffering. London: Verso. Fitzpatrick, Matthew. 2008. The Pre-History of the Holocaust? The Sonderweg and Historikerstreit Debates and the Abject Colonial Past. Central European History, 41, 477-503. Fitzmaurice, Andrew. 2008. Anticolonialism in Western Political Thought: The Colonial Origins on the Concept of Genocide. In Moses (ed.) Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in World History. New York, NY: Berghahn Books, 55-80. Fletcher, Jonathan. 1997. Violence and Civilization: An Introduction to the Work of Norber Elias. Cambridge: Polity. Fletcher, Luke. 2007. Turning Interahamwe: Individual and Community Choices in the Rwandan Genocide. Journal of Genocide Research, 9 (1), 25-48. Foucault, Michel. 1977. Discipline and Punish: The Birth of the Prison, trans. Sheridan. New York, NY: Vintage. Foucault, Michel. 1982. On the Genealogy of Ethics: An Overview of a Work in Progress. In Dreyfus & Rabinow (eds.) Michel Foucault: Beyond Structuralism and Hermeneutics. Chicago, IL: University of Chicago Press, 229-252. Foucault, Michel. 2003. Society Must Be Defended: Lectures at the Collège de France, 1975-76, eds. & trans. Bertani & Fontana. New York, NY: Picador. Fraser, Nancy. 2000. Rethinking Recognition. New Left Review, 6 (1), 107-120. Freeman, Jim. 1974. Ritual and History: The Case of Nyabingi. Cahiers d’Études Africaines, 14 (53), 170-180. 277

Frère, Marie-Soleil. 2007. The Media and Conflicts in Central Africa. London: Lynn Reinner Publishers. Friedrichs, Jörg & Kratochwil, Friedrich. 2009. On Acting and Knowing: How Pragmatism Can Advance International Relations Research and Methodology. International Organization, 63 (4), 701-731. Friese, Heidrun & Wagner, Peter. 2000. When ‘the Light of the Great Cultural Problems Moves on’: On the Possibility of a Cultural Theory of Modernity. Thesis Eleven, 61 (1), 25-40. Fujii, Lee Ann. 2004. Transforming the Moral Landscape: The Diffusion of a Genocidal Norm in Rwanda. Journal of Genocide Research, 6 (1), 99-114. Fukuyama, Francis, 1989. The end of history? The National Interest, (16), 3-18. Gadamer, Hans-Georg. 2004 [1975]. Truth and Method, trans. Weinsheimer & Marshall (2nd ed.). New York, NY: Continuum. Gadamer, Hans-Georg. 2006. Classical and Philosophical Hermeneutics, trans. Palmer. Theory, Culture & Society, 23 (1), 29-56. Gadamer, Hans-Georg. 2008 [1976]. Philosophical Hermeneutics, trans. & ed. Linge. Berkeley, CA: University of California Press. Gahama, Joseph. 2001. Le Burundi sous administration belge: Le période du mandate 1919-1939. Paris: Karthala. Gakuba, Laurent. 1991. Rwanda, 1931-1959: Derniers decennies de la monarchie sous la colonisation belge. Paris: La Pensée Universelle. Gann, Lewis. 1979. The Rulers of Belgian Africa, 1884-1914. Princeton, NJ: Princeton University Press. Gatwa, Tharcisse. 2005. The Churches and Ethnic Ideology in the Rwandan Crises, 1900-1994. Milton Keynes: Paternoster. Gellately, Robert & Kiernan, Ben (eds.). 2003. The Spectre of Genocide: Mass Murder in Historical Perspective. Cambridge: Cambridge University Press. Gerwath, Robert & Malinowki, Stephan. 2009. Hannah Arendt’s Ghosts: Reflections on the Disputable Path from Windhoek to Auschwitz. Central European History, 42, 279-300. Geyer, Dietrich. 1987. Russian Imperialism: The Interaction of Domestic and Foreign Policy, 1860- 1914. New Haven, CT: Yale University Press. Giddens, Anthony. 1971. Capitalism and Modern Social Theory: An Analysis of the Writings of Marx, Durkheim and Max Weber. Cambridge: Cambridge University Press. Giddens, Anthony. 1990. Consequences of Modernity. Cambridge: Polity. Gidley, Ben. 2004. Doing Historical and Archival Research. In Seale (ed.) Researching Society and Culture (second edition). London: SAGE, 249-265. Gilroy, Paul. 2005. Postcolonial Melancholia. New York, NY: Columbia University Press. Go, Julian. 2013. For a Postcolonial Sociology. Theory and Society, 42 (1), 25-55. Go, Julian (ed.) 2013a. Postcolonal Sociology (special issue). Political Theory and Social Theory, 24 (1). Goddeeris, Idesbald. 2015. Postcolonial Belgium. Interventions: International Journal of Postcolonial Studies, 17 (3), 434-451. Gordon, Lewis. 1996. Fanon’s Tragic Revolutionary Violence. In Gordon (ed.) Fanon: A Critical Reader. Oxford: Blackwell, 297-308. Gourevitch, Philip. 1998. We wish to inform you that tomorrow we will be killed with our families. London: Picador. Gramsci, Antonio. 1971. Selections from the Prison Notebooks, trans. & ed. Hoare & Nowell-Smith. London: Lawrence & Wishart. Greene, Sandra. 2003. The Multiple Worlds of African Studies. In Mirsepassi, Basu & Stirton (eds.) Localizing Knowledge in a Globalizing World: Recasting the Area Studies Debate. Syracuse, NY: Syracuse University Press, 137-147. 278

Greenland, Jeremy. 1976. Ethnic Discrimination in Burundi and Rwanda. In Veenhoven (ed.) Case Studies in Huan Rights and Fundamental Freedoms: A World Survey. The Hague: Martnius Nijhoff, 98-133. Griffin, Roger. 2007. Modernism and Fascism: The Sense of a Beginning under Mussolini and Hitler. Basingstoke: Palgrave Macmillan. Grondin, Jean. 1994. Introduction to Philosophical Hermeneutics, trans. Joel Weinsheimer. New Haven, CT: Yale University Press. Gros, Jean-Germain. 1996. Towards a Taxomony of Failed States in the New World Order: Decaying Somalia, Liberia, Rwanda and Haiti. Third World Quaterly, 17 (3), 455-472. Habermas, Jürgen. 1981. Modernity versus Postmodernity. New German Critique, 22 (4),3-14. Habermas, Jürgen. 1984. The Theory of Communicative Action – Volume 1: Reason and the Rationalization of Society. Boston, MA: Beacon Press. Habermas, Jürgen. 1987. The Theory of Communicative Action – Volume 2: Lifeworld and System: A Critique of Functionalist Reason. Boston, MA: Beacon Press. Habermas, Jürgen. 1987a. The Philosophical Discourse of Modernity: Twelve Lectures, trans. Fredrick Lawrence. Cambridge: Polity. Habermas, Jürgen. 1988. On the Logic of the Social Sciences, trans Nicholsen and Stark. Cambridge, MA: MIT Press. Habermas, Jürgen. 1996. Modernity: An Unfinished Project. In d’Entrèves & Benabib (eds.) Habermas and the Unfinished Project of Modernity: Critical Essays on the Philosophical Discourse of Modernity. Cambridge, MA: MIT Press. Hall, Peter & Tarrow, Sidney. 2001. Globalization and Area Studies: When is too Broad too Narrow? In O’Meara, Mehlinger & Newman (eds.) Changing Perspectives on International Education. Bloomington, IN: Indiana University Press, 96-100. Hall, Stuart. 1992. The West and the Rest: Discourse and Power. In Hall & Gieben (eds.) Formations of Modernity. Oxford: Blackwell, 275-332. Hallward, Peter. 2001. Absolutely Postcolonial: Writing Between the Singular and the Specific. Manchester: Manchester University Press. Hammersley, Martyn. 2011. Methodology: Who Needs It? London: SAGE. Hansen, Peo & Jonsson, Stefan. 2014. Eurafrica: The Untold History of European Integration and Colonialism. London: Bloomsbury. Harding, Sandra (ed.). 2004. The Feminist Standpoint Theory Reader: Intellectual and Political Controversies. London: Routledge. Harff, Barbara. 2003. No Lessons Learned from the Holocaust? Assessing Risks of Genocide and Political Mass Murder since 1955. American Political Science Review, 97, 57-73. Harrington, Austin. 2000. Alfred Schutz and the ‘Objectifying Attitude’. Sociology, 34 (4), 727-740. Harrington, Austin. 2001. Hermeneutic Dialogue and Social Science: A Critique of Gadamer and Habermas. London: Routledge. Harrington, Austin. 2016. German Cosmopolitan Social Thought and the Idea of the West: Voices from Weimar. Cambridge: Cambridge University Press. Harrison, David. 1988. The Sociology of Modernization and Development. London: Routledge. Hartz, Louis. 1969. The Founding of New Societies: Studies in the History of the United States, Latin America, South Africa, Canada, and Australia. Orlando, FL: Mariner Books. Harvey, David. 2005. A Brief History of Neoliberalism. Oxford: Oxford University Press. Hatzfield, Jean. 2003. Machete Season: The Killers in Rwanda Speak. New York, NH: Farrar, Straus and Giroux. Hegel, G. W. F. 2001 [1837]. The Philosophy of History, trans. J. Sibree. Kitchener: Batoche Books. Heidegger, Martin. 1996 [1953]. Being and Time, trans. Stimbaugh. Albany, NY: SUNY Press. 279

Heidegger, Martin. 2003 [1938]. The Age of the World Picture. In Delanty and Strydom (eds.) Philosophies of Social Science: The Classic and Contemporary Readings. Berkshire: Open University Press, 147-151. Held, David & McGrew, Anthony. 2003. The Great Globalization Debate: An Introduction. In Held & McGrew (eds.) The Global Transformations Reader: An Introduction to the Globalization Debate (2nd ed.). Cambridge: Polity, 1-50. Heller, Agnes. 2005. The Three Logics of Modernity and the Double Bind of the Imagination. Thesis Eleven, 81 (1), 63-79. Heller, Agnes. 2010. Radical Evil in Modernity: On Genocide, Totalitarian Terror and the Holocaust. Thesis Eleven, 101, 106-117. Herman, Edward & Peterson, David. 2010. The Politics of Genocide. New York: Monthly Review Press. Hilberg, Raul. 1985 [1961]. The Destruction of the European Jews. Teaneck, N.J: Holmes and Meier Publishers. Hintjens, Helen. 2001. When Identity Becomes a Knife: Reflecting on the Genocide in Rwanda. Ethnicities, 1 (1), 25-55. Hintjens, Helen. 2008. Rwanda and Burundi. In Herb & Kaplan (eds.) Nations and Nationalism: A Global Historical Overview, 4 vols. Oxford: ABC-Clio, 1669-1682. Hinton, A. L. 2002. The Dark Side of Modernity: Toward an Anthropology of Modernity. In Hinton (ed.) Annihilating Difference: The Anthropology of Genocide. Berkeley, CA: University of California Press. Hinton, A. L. 2008. Savages, Subjects, and Sovereigns: Conjunctions of Modernity, Genocide, and Colonialism. In Moses (ed.) Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in World History. New York, NY: Berghahn Books, 440-460. Hinton, A. L. 2012. Critical Genocide Studies. Genocide Studies and Prevention, 7 (1), 4-15. Hinton, A. L., La Pointe, T. & Irvin-Erickson, D. (eds.). 2013. Hidden Genocides: Power, Knowledge, Memory. New Brunswick, NJ: Rutgers University Press. Hitchens, Christopher. 2008. God is not Great: How Religion Poisons Everything. McClelland & Stewart Hobsbawm, Eric & Ranger, Terence. 1983. The Invention of Tradition. Cambridge: Cambridge University Press. Hochschild, Adam. 1999. King Leopold’s Ghost: A Story of Greed, Terror, and Heroism in Colonial Africa. Basingstoke: Panmacmillan. Holmwood, John. 2010. Sociology’s Misfortune: Disciplines, Interdisciplinarity and the Impact of Audit Culture. British Journal of Sociology, 61 (4), 639-658. Human Rights Watch. March, 1993. Report of the International Commission of Investigation on Human Rights Violations in Rwanda since October 1, 1990. Human Rights Watch. January, 1994. Arming Rwanda: The Arms Trade and Human Rights Abuses in the Rwandan War. Ichijo, Atsuko. 2013. Nationalism and Multiple Modernities: Europe and Beyond. Basingstoke: Palgrave. ICTR (International Criminal Tribunal for Rwanda). 2008, December 2nd. THE PROSECUTOR v. Simon BIKINDI Case No. ICTR-01-72-T. Iliffe, John. 1969. Tanganyika Under German Rule, 1905-1912. Cambridge: Cambridge University Press. Inglis, David. 2013. What is Worth Defending in Sociology Today? Presentism, Historical Vision and the Uses of Sociology. Cultural Sociology, 8 (1), 99-118. International Crisis Group. 2016, September 28th. The African Union and the Burundi Crisis: Ambition versus Reality. Nairobi/Brussels: Crisis Group Africa Briefing N ̊ 22. 280

Inwood, Michael. 2000. Martin Heidegger: A Very Short Introduction. Oxford: Oxford University Press. Irish Times, 2016, November 9. Far-right congratulates Trump: ‘Their World is collapsing. Ours is being built’. Irish Times. Available online at: http://www.irishtimes.com/news/world/us/far- right-congratulates-trump-their-world-is-collapsing-ours-is-being-built-1.2861153 Jack, E. M. 1914. On the Congo Frontier: Exploration and Sport. London: T. Fisher Unwin. James, C. L. R. 1989 [1938]. The Black Jacobins: Toussaint L’Ouverture and the San Domingo Revolution (2nd ed.) New York, NY: Vintage. Jaspers, Karl. 1953 [1949]. The Origin and Goal of History, trans. Bullock. New Haven, CT: Yale University Press. Jefremovas, Villia. 1997. Power and Ethnicity, Ethnography and History in Rwanda. Anthropoligica, 39 (1), 91-104. Jefremovas, Villia. 2002. Brickyards to Graveyards: From Production to Genocide in Rwanda. New York, NY: SUNY Press. Jewsiewicki, Bogumil. 1986. Collective Memory and the Stakes of Power: A Reading of Popular Zairian Historical Discourses. History in Africa, 13 (1), 195-223. Joas, Hans. 1996. The Creativity of Action, trans. Gaines & Keast. Cambridge: Polity. Joas, Hans. 1998. Bauman in Germany: Modern Violence and the Problems of German Self- Understanding. Theory, Culture & Society, 15 (1), 47-55. Joas, Hans. 2000. The Genesis of Values, trans. Moore. Chicago, IL: University of Chicago Press. Joas, Hans. 2000a. Comment on Göran Therborn. In Schelke et al (eds.) Paradigms of Social Change: Modernization, Development, Transformation, Evolution. Frankfurt/New York, NY: Campus Verlag, 73-77. Joas, Hans. 2003. War and Modernity: Studies in the History of Violence in the 20th Century, trans. Livingstone. Cambridge: Polity. Joas, Hans. 2013. The Sacredness of the Person: A New Genealogy of Human Rights, trans. Skinner. Washington DC, MD: Georgetown University Press. Joas, Hans & Knöbl, Wolfgang. 2009. Social Theory: Twenty Introductory Lectures trans. Skinner. Cambridge: Cambridge University Press. Joas, Hans & Knöbl, Wolfgang. 2013. War in Social Thought: Hobbes to the Present, trans. Skinner. Princeton, NJ: Princeton University Press. Joas, Hans; Sennett, Richard & Antje Gimmler. 2006. Creativity, Pragmatism, and the Social Sciences. Distinktion: Scandinavian Journal of Social Theory, 7 (2), 5-31. Jones, Adam. 2009. On the Genocidal Aspect of Certain Subaltern Uprisings: A Research Note. In Jones & Robins (eds.) Genocides by the Oppressed: Subaltern Genocide in Theory and Practice. Bloomington, IN: Indiana University Press, 47-57. Jones, Adam. 2011. Genocide: A Comprehensive Introduction, 2nd ed. London: Routledge. Jones, Adam. 2014. The Great Lakes Genocides: Hidden Histories, Hidden Precedents. In Hinton, La Pointe & Irvin-Erickson (eds.) Hidden Genocides: Power, Knowledge, Memory. New Brunswick, NJ: Rutgers University Press, 129-148. Jones, Adam & Robins, Nicholas (eds.). 2009. Genocides by the Oppressed: Subaltern Genocide in Theory and Practice. Bloomington, IN: Indiana University Press. Kaczmarska, Katarzyna. 2016. ‘But in Asia we too are Europeans’: Russia’s Multifaceted Engagement with the Standard of Civilisation. International Relations, 30 (4), 432-455. Kagame, Alexis. 1945. Le Rwanda et son roi. Aequatoria, 8 (2), 41-58. Kagame, Alexis. 1951. La poésie dynastique au Rwanda. Brussels: Institute Royal Colonial Belge. Kaldor, Mary. 2007. New and Old Wars: Organized Violence in a Global Age (2nd Edition). Cambridge: Polity. Kamatali, Jean-Marie. 2013. State-Building in Rwanda. In Keih Jr. & Agbese (eds.) Reconstructing the Authoriatarian State in Africa. London, Routledge, 154-179. 281

Kamola, Isaac. 2007. The Global Coffee Economy and the Production of Genocide in Rwanda. Third World Quarterly, 28 (3), 571-592. Kaplan, Robert. 2002. The Coming Anarchy: Shattering the Dreams of the Post Cold War World. New York, NY: Vintage. Katz, Steven. 2001. The Uniqueness of the Holocaust: The Historical Dimension. In Rosenbaum (ed.) Is the Holocaust Unique? Perspectives on Comparative Genocide. Boulder, CO: Westview, 49-68. Kaya, Ibrahim. 2004. Social Theory and Later Modernities: The Turkish Experience. Liverpool: Liverpool University Pres. Kayoya, Michael, 1971. Entre deux mondes: sur la route du développement. Bujumbura: Éditions des Presses Lavigerie. Keane, Fergal. 1995. Season of Blood: A Rwandan Journey. London: Viking. Kerényi, Karl. 2015 [1944]. Hermes: Guide of Souls, trans. Stein. Thompson, CT: Spring Publications. Kern, Thomas; Mayer, Lotta, & Nam, Sang-hui. 2014. The Construction of Regional Identities in East Asia. In Arjomand (ed.) Social Theory and Regional Studies in the Global Age. Albany, NY: SUNY Press, 415-436. Kiernan, Ben. 2009. Genocide and Resistance in Southeast Asia: Documentation, Denial and Justice in Cambodia and East Timor. New Brunswick, NJ: Transaction Publishers. Kilminster, Richard. 2007. Norbert Elias: Post-Philosophical Sociology. London: Routledge. Kilminster, Richard. 2013. Critique and Overcritique in Sociology. Human Figurations, 2 (2), available online at: http://hdl.handle.net/2027/spo.11217607.0002.205 Kinsella, Elisabeth Anne. 2006. Hermeneutics and Critical Hermeneutics: Exploring Possibilities within the Art of Interpretation. Forum Qualitative Sozialforschung, 7 (3), available online at: http://www.qualitativeresearch.net/index.php/fqs/article/view/145/319#g22 Kissi, Edward. 2006. Revolution and Genocide in Ethiopia and Cambodia. Lanham, MD: Lexington Books. Kocka, Jürgen. 2003. Comparison and Beyond. History and Theory, 42 (1), 39-44. Kögler, Hans-Herbert. 2006. Hermeneutics, Phenomenology, and Philosophical Anthropology. In Delanty (ed.) Handbook of Contemporary European Social Theory. London: Routledge, 203- 217. Kołakowski, Leszek. 1990. Modernity on Endless Trial. Chicago, IL: University of Chicago Press. Koselleck, Reinhart. 2004 [1979]. Futures Past: On the Semantics of Historical Time, trans. Tribe. New York, NY: Columbia University Press. Knöbl, Wolfgang. 2002. Modernization Theory, Modernization & African Modernities: An Outsider’s View. In Deutsche, Probst & Schmidt (eds.) African Modernities: Entangled Meanings in Current Debate. Oxford: James Currey Ltd, 158-178. Kressel, Neil. 1996. Mass Hate: The Global Rise of Genocide and Terror. New York, NY: Plenum Press. Kühne, Thomas. 2013. Colonialism and the Holocaust: Continuities, Causations, and Complexities. Journal of Genocide Research, 15 (3), 339-362. Kuper, Leo. 1970. Continuities and Discontinuities in Race Relations: Evolutionary or Revolutionary Change. Cahiers d’Étude Africaines, 10 (39), 361-383. Kuper, Leo. 1971. Theories of Revolution and Race Relations. Comparative Studies in Society and History, 13 (1), 87-107. Kuper, Leo. 1982. Genocide: Its Political Use in the Twentieth Century. New Haven, CT: Yale University Press. Kuperman, Alan. 2004. Provoking Genocide: A Revised History of the Rwandan Patriotic Front. Journal of Genocide Research, 6 (1), 61-84. 282

Kurasawa, Fuyuki. 2000. The Ethnological Counter-Current in Sociology. International Sociology, 15 (1), 11-31. Kurasawa, Fuyuki. 2004. The Ethnological Imagination: A Cross-Cultural Critique of Modernity. Minneapolis, MN: University of Minnesota Press. Laclau, Ernesto. 1990. New Reflections on the Revolution of our Time. London: Verso. Laely, Thomas. 1997. Peasants, Local Communities, and Central Power in Burundi. The Journal of Modern African Studies, 35 (4), 695-716. Lal, Vinay. 2007. The Concentration Camp and Development: the Pasts and Future of Genocide. In Moses & Stone (eds.) Colonialism and Genocide. London: Routledge, 124-147. Latham-Koenig, A. L. 1962. Ruanda-Urundi on the Threshold of Independence. The World Today, 18 (7), 288-295. Latour, Bruno. 1993. We Have Never Been Modern, trans. Porter. Cambridge, MA: Harvard University Press. Law, Ian. 2013. Racism and Ethnicity: Global Debates, Dilemmas, Directions. London: Routledge. Lazarus, Neil. 2004. The Fetish of ‘the West’ in Postcolonial Theory. In Bartolovich & Lazarus (eds.) Marxism, Modernity and Postcolonial Studies. Cambridge: Cambridge University Press, 43-64. Lazzarato, Mauricio. 2011. The Making of the Indebted Man. Los Angeles, CA: Semiotext(e). Lemarchand, René. 1966. Social Change and Political Modernisation in Burundi. The Journal of Modern African Studies, 4 (4), 401-433. Lemarchand, René. 1970. Rwanda and Burundi. Westport, CT: Praeger. Lemarchand, René. 1977. Rwanda. In Lemarchand (ed.) African Kingships in Perspective: Political Change and Modernization in Monarchical Settings. London: Frank Cass & Co. Ltd, 67-92. Lemarchand, René. 1977a. Burundi. In Lemarchand (ed.) African Kingships in Perspective: Political Change and Modernization in Monarchical Settings. London: Frank Cass & Co. Ltd, 93-128. Lemarchand, René. 1980. Ethnic Genocide. Issue: A Journal of Opinion. 10 (1-2), 54-61. Lemarchand, René. 1994. Burundi: Ethnocide as Discourse and Practice. Cambridge: Cambridge University Press. Lemarchand, René. 1995. Rwanda: The Rationality of Genocide. Issue: A Journal of Opinion, 23 (2), 8-11. Lemarchand, René. 2009. The Dynamics of Violence in Central Africa. Philadelphia, PA: University of Pennsylvania Press Lemarchand, René (ed.). 2011. Forgotten Genocides: Oblivion, Denial and Memory. Philadelphia, PA: University of Pennsylvania Press. Lemarchand, René. 2011a. Burundi 1972: Genocide Denied, Revised, and Remembered. In Lemarchand (ed.) Forgotten Genocides: Oblivion, Denial, Memory. Philadelphia, PA: University of Pennsylvania Press, 37-50. Lemarchand, René and Martin, Peter. 1974. Selective Genocide in Burundi. London: Minority Rights Group. Lemarchand, René and Reyntjens, Filip. 2011. Mass Murder in Eastern Congo, 1996-1997. In Lemarchand (ed.) Forgotten Genocides: Oblivion, Denial, Memory. Philadelphia, PA: University of Pennsylvania Press, 20-36. Lemkin, Raphael. 2009 [1945]. Genocide: A Modern Crime. In Totten & Bartrop (eds.) The Genocide Studies Reader. London: Routledge, 6-11. Leurquin, Philippe. 1963. Agricultural Change in Ruanda-Urundi: 1945-1960. Food Research Institute Studies, 4 (1), 39-89. Levene, Mark. 2005. Genocide in the Age of the Nation-State. Vol. 1, The Meaning of Genocide. London: IB Tauris. Levene, Mark. 2005a. Genocide in the Age of the Nation-State. Vol. 2, The Rise of the West and the Coming of Genocide. London: IB Tauris. 283

Levene, Mark. 2008. Why is the Twentieth Century the Century of Genocide? Journal of World History, 11 (2), 305-336. Levine, Donald. 2000 [1974]. Greater Ethiopia: The Evolution of a Multi-Ethnic Society, 2nd ed. Chicago: University of Chicago Press. Lindner, Evelin. 2009. Genocide, Humiliation, and Inferiority: An Interdisciplinary Perspective. In Jones & Robins (eds.) Genocides by the Oppressed: Subaltern Genocide in Theory and Practice. Bloomington, IN: Indiana University Press, 138-158. Lieberman, Benjamin. 2012. From Definition to Process: the Effects and Roots of Genocide. In Jones (ed.) New Directions in Genocide Research. New York, NY: Routledge, 3-17. Linden, Ian. 1977. Church and Revolution in Rwanda. Manchester: Manchester University Press. Lindqvist, Sven. 2012. Saharan Journey: Exterminate All the Brutes and Desert Divers. London: Granta Books. Livingstone, David. 1865. Narrative of an Expedition to the Zambesi and its Tributaries; and of the Discovery of the Lakes Shirwa and Nyassa, 1858-1864. London: John Murray. Locke, Karen; Golden-Biddle, Karen & Feldman, Martha. 2008. Making Doubt Generative: Rethinking the Role of Doubt in the Research Process. Organization Science, 19 (6), 907- 918. Loft, Francis. 1988. Background to the Massacres in Burundi. Review of African Political Economy, 43 (1), 88-93 Long, Katy. 2012. Rwanda’s First Refugees: Tutsi Exile and International Response. Journal of Eastern African Studies, 6 (2), 211-229. Longman, Timothy. 2001. Church Politics and the Genocide in Rwanda. Journal of Religion in Africa, 31 (2), 163-186. Longman, Timothy. 2004. Placing Genocide in Context: Research Priorities for the Rwanda Genocide. Journal of Genocide Research, 6(1), 29-45. Longman, Timothy. 2011. Christianity and Genocide in Rwanda. Cambridge: Cambridge University Press. Louis, W. R. 1963. Ruanda-Urundi, 1884-1919. Oxford: Clarendon Press. Luft, Aliza. 2015. Towards a Dynamic Theory of Action at the Micro Level of Genocide: Killing, Desistance, and Saving in 1994 Rwanda. Sociological Theory, 33 (2), 148-172. Lyotard, Jean-Francois. 1984. [1979] The Postmodern Condition. Minneapolis, MN: University of Minnesota Press. Macionis, John & Plummer, Ken. 2008. Sociology: A Global Introduction (4th ed.). Essex: Prentice Hall. Mafeje, Archie. 1991. The Theory and Ethnography of African Social Formations: The Case of the Interlacustrine Kingdoms. London: CODESRIA. Magnarella, Paul. 2005. How could it happen? The Background and Causes of the Genocide in Rwanda. Journal of International Criminal Justice, 3 (1), 801-822. Magubane, Zine. 2005. Overlapping Territories and Intertwined Histories: Historical Sociology’s Global Imagination. In Adams, Clemens & Orloff, Remaking Modernity: Politics, History, and Sociology. Durham, NC: Duke University Press, 92-108. Mahoney, James. 2000. Path Dependence in Historical Sociology. Theory and Society, 29 (4), 507- 548. Maldonado-Torres, Nelson. 2007. On the Coloniality of Being. Cultural Studies, 21(2), 240-270. Malešević, Siniša. 2010. The Sociology of War and Violence. Cambridge: Cambridge University Press. Mamdani, Mahmood. 1996. Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton, NJ: Princeton University Press. Mamdani, Mahmood. 2001. When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda. Princeton, NJ: Princeton University Press. 284

Mamdani, Mahmood. 2011, January 13th. The Invention of the Indigène. Pambazuka News, Issue 512. Available online at http://www.pambazuka.org/en/category/features/70061 Mamdani, Mahmood. 2012. Define and Rule: Native as Political Identity. Cambridge, MA: Harvard University Press. Mann, Michael. 2004. The Dark Side of Democracy: Explaining Ethnic Cleansing. Cambridge: Cambridge University Press. Mantz, Jeffrey. 2008. Improvisational economies: Coltan production in eastern Congo. Social Anthropology, 16 (1), 34-50. Maquet, Jacques. 1954. Le système des relations sociales dans le Rwanda ancient. Tervuren. Musée Royal de l’Afrique Central. Maquet, Jacques. 1961. The Premise of Inequality in Ruanda: A Study of Political Relations in a Central African Kingdom. Oxford: Oxford University Press. Maquet, Jacques & D’Hertefelt, Marcel. 1959. Élections en société féodale: Une étude sur l’introduction du vote populaire au Ruanda-Urundi. Brussels: Académie royale des sciences coloniales. Marcuse, Herbert. 1991 [1964]. One-Dimensional Man: Studies in the Ideology of Advanced Industrial Society. London: Routledge. Marima, Tendai. 2015, December 6th. Escaping Burundi: Life in Refuge. Al Jazeera. Available online at http://www.aljazeera.com/indepth/inpictures/2015/12/escaping-burundi-life-refuge- 151202060334408.html Mariro, Augustin. 2005. Burundi: La 1ère crise ethnique: genèse et context géopolitique. Paris: L’Harmattan. Marshman, Sophia. 2008. Bauman on Genocide – Modernity and Mass Murder: From Classification to Annihilation? In Jacobsen and Poder (eds.) The Sociology of Zygmunt Bauman: Challenges and Critique. Aldershot: Ashgate: 75-94. Marx, Karl. 1852. The Eighteenth Brumaire of Louis Bonaparte. Available online at: https://www.marxists.org/archive/marx/works/1852/18th-brumaire/ch01.htm Marx, Karl & Engels, Friedrich. 2004 [1848]. The Communist Manifesto. London: Penguin. Mazower, Mark. 2006. An International Civilization? Empire, Internationalism, and the Crisis of the Mid-Twentieth Century. International Affairs, 82 (3), 553-566. Mbembe, Achille. 2001. On the Postcolony. Los Angeles: University of California Press. Mbembe, Achille & Nuttall, Sarah. 2004. Writing the World from an African Metropolis. Public Culture, 16 (3), 347-372. McDoom, Omar. 2010. War and Genocide in Africa’s Great Lakes Since Independence. In Bloxham & Moses (eds.) The Oxford Handbook of Genocide Studies. Oxford: Oxford University Press, 550-578. McLean, Ruth & Burke, Jason. 2016, November 10th. Democratic Republic of the Congo ‘faces civil war’ if president fails to quit. The Guardian. Available online at: https://www.theguardian.com/world/2016/nov/10/democratic-republic-of-the-congo-faces- civil-war-if-president-fails-to-quit McLennan, Gregor. 2010. Eurocentrism, Sociology, Secularity. In Gutiérrez Rodríguez, Boatcă & Costa (eds.) Decolonizing European Sociology: Transdisciplinary Approaches. Farnham: Ashgate. McNeish, Hannah. 2013. CAR: The broken heart of Africa. Al Jazeera. Available online at http://www.aljazeera.com/indepth/features/2013/11/car-broken-heart-africa 20131123101734112996.html McNulty, Mel. 1999. Media Ethnicization and the International Response to War and Genocide in Rwanda. In Allen, T & Seaton, J (Eds), The Media of Conflict. London: Zed Books Ltd, 268- 287. 285

Meister, Robert. 2012. After Evil: A Politics of Human Rights. New York, NY: Colombia University Press. Melady, Thomas. 1974. Burundi: the Tragic Years. Maryknoll, NY: Orbis Books. Melber, Henning & Jones, John (eds.). 2008. Special Issue: Revisiting the Heart of Darkness – Explorations into Genocide and other forms of Mass Violence. Development Dialogue, 50 (1). Melson, Robert. 1992. Revolution and Genocide: On the Origins of the Armenian Genocide and the Holocaust. Chicago, IL: University of Chicago Press. Melvern, Linda. 2006. Conspiracy to Murder: The Rwandan Genocide. London: Verso. Mignolo, Walter. 2001. Coloniality at Large: The Western Hemisphere in the Colonial Horizon of Modernity. The New Centennial Review, 1 (2), 19-54. Mignolo, Walter. 2011. The Darker Side of Western Modernity: Global Futures, Decolonial Options. Durham, NC: Duke University Press. Mignolo, Walter. 2012. Local Histories/Global Designs: Coloniality, Subaltern Knowledges, and Border Thinking. Princeton, NJ: Princeton University Press. Mills, C. W. 2000 [1959]. The Sociological Imagination. Oxford: Oxford University Press. Ministère des colonies. 1926. Rapport sur l’Administration Belge au Ruanda-Urundi, 1925. Bruxelles: Administration Belge du Ruanda-Urundi. Ministère des colonies. 1933. Rapport sur l’Administration belge du Ruanda-Urundi pendant l’annee 1933. Bruxelles: Administration Belge du Ruanda-Urundi. Mogalakwe, Monageng. 2009. The Documentary Research Method – Using Documentary Sources in Social Research. Eastern African Social Science Research Review, 25 (1), 43-58. Moore, J. E. S. 1901. To the Mountains of the Moon: Being an Account of the Modern Aspect of Central Africa, and of Some Little Known Regions Traversed by the Tanganyika Expedition, in 1899 and 1900. London: Hurst & Blackett Ltd.. Moore, W. E. 1966. Global Sociology: The World as a Singular System. The American Journal of Sociology, 71 (5), 575-482. Moran, Dermot. 2000. Introduction to Phenomenology. London: Routledge. Moses, A. D. 2002. Conceptual Blockages and Definitional Dilemmas in the ‘Racial Century’: Genocides of Indigenous Peoples and the Holocaust. Patterns of Prejudice, 36 (4), 7-36. Moses, A. D. 2008. Empire, Colony, Genocide: Keywords in the Philosophy of History. In Moses (ed.) Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in World History. New York, NY: Berghahn Books, 3-54. Moses, A. D. 2008a. Genocide and Modernity. In Stone (ed.) The Historiography of Genocide. Basingstoke: Palgrave Macmillan, 156-193. Moses, A. D. 2008b. Toward a Theory of Critical Genocide Studies. Online Encyclopedia of Mass Violence [online]. Accessed 25 November 2013, URL : http://www.massviolence.org/Toward-a-Theory-of-Critical-Genocide-Studies, ISSN 1961- 9898 Moses, A. D. 2011. Genocide and the Terror of History. Parallax, 17 (4), 90-108. Moses, A. D. 2011a. Paranoia and Partisanship: Genocide Studies, Holocaust Historiography, and the ‘Apocalyptic Conjuncture’. The Historical Journal, 54 (2), 553-583. Mudimbe, V. Y. 1988. The Invention of Africa: Gnosis, Philosophy and the Order of Knowledge. London: James Curry. Mworoha, Émile. 1986. Politiques et actions du gouvernement du Burundi en matière de développement rural. Revue Tiers Monde, 27 (106), 339-355. Mworoha, Émile. 1987. Histoire du Burundi: Des origines à la fin du XIXe siècle. Paris: Hatier. Mworoha, Émile & Mukuri, Melchior. 2004. Problématique de la périodisation historique pour la région des Grands Lacs. Afrique & histoire, 1 (2), 67-83. Mukerjee, Radhakamal. 1926. Regional Sociology. New York, NY and London: The Century Co. 286

Myers, R. H. & Peattie, Mark (eds.). 1984. The Japanese Colonial Empire, 1895-1945. Princeton, NJ: Princeton University Press. Nagel, Joane. 1994. Constructing Ethnicity: Creating and Recreating Ethnic Identity and Culture. Social Problems, 41 (1), 152-176. Nahimana, Salvator & Hatungimana (eds.), 2004. Melchior Ndadaye: pour le Burundi nouveau: campagne 1992-1993. Paris: L’Harmattan. Ndahinda, Felix M. 2011. Indigenousness in Africa: A Contested Legal Framework for Empowerment of ‘Marginalized’ Communities. The Hague: TMC Asser Press. Ndarishikanye, Barnabé. 1999. Burundi: des identités ethnico-politiques forgées dans la violence. Canadian Journal of African Studies / Revue canadienne des études africaines, 33(2), 231- 291. Ndikumana, Léonce. 1998. Institutional Failure and Ethnic Conflicts in Burundi. African Studies Review, 41 (1), 29-47. Ndlovu-Gatsheni, Sabelo. 2013. Empire, Global Coloniality, and African Subjectivity. New York, NY: Berghahn. Nelson, Jack. 1983. The ‘Premise of Inequality’ in Rwanda: A critique. Nenon, Thomas. 2016. Horizonality. In Keane & Lawn (eds.) The Blackwell Companion to Hermeneutics. Oxford: Wiley-Blackwell. Nest, Michael. 2011. Coltan. Cambridge: Polity. Newbury, Catherine. 1980. Ubureetwa and Thangata: Catalysts to Peasant Political Consciousnes in Rwanda and Malawi. Canadian Journal of African Studies, 14 (1), 97-111. Newbury, Catherine. 1983. Colonialism, Ethnicity, and Rural Political Protest: Rwanda and Zanzibar in Comparative Perspective. Comparative Politics, 15 (3), 253-280. Newbury, Catherine. 1988. The Cohesion of Oppression: Clientship and Ethnicity in Rwanda, 1860- 1960. New York, NY: Columbia University Press. Newbury, Catherine. 1998. Ethnicity and the Politics of History in Rwanda. Africa Today, 45 (1), 7- 24. Newbury, David. 1980. Lake Kivu Regional Trade in the Nineteenth Century. Journal des Africanistes, 50 (2), 7-30. Newbury, David. 1997. Irredentist Rwanda: Ethnic and Territorial Frontiers in Central Africa. Africa Today, 44 (2), 211-222. Newbury, David. 1998. Understanding Genocide. African Studies Review, 41 (1), 73-97. Newbury, David. 2001. Precolonial Burundi and Rwanda: Local Loyalties, Regional Royalties. The International Journal of African Historical Studies, 34 (2), 255-314. Newbury, David. 2005. Returning Refugees: Four Historical Patterns of ‘Coming Home’ to Rwanda. Comparative Studies in Society and History, 47 (2), 252-285. Newbury, David. 2011. Editor’s Introduction: Situating the Rwandan Court at the Time of Musinga’s Accession to Power. In Des Forges (auth.) Defeat is the Only Bad News: Rwanda under Musinga, 1896-1931. Madison, WI: University of Wisconsin Press, xxiii – xxvii. Newbury, David. 2012. Canonical Conventions in Rwanda: Four Myths of Recent Historiography in Central Africa. History in Africa, 39 (1), 41-76. Newbury, David & Newbury, Catherine. 2000. Bringing the Peasants Back In: Agrarian Themes in the Construction and Corrosion of Statist Historiography in Rwanda. The American Historical Review, 105 (3), 832-877. Ngabirano, Maximiano & Tsimbi, David. 2012. Memory and Historical Realities in Shaping Politics and Armed Conflicts in Rwanda. In Ewusi (ed.) Weaving Peace: Essays on Peace, Governance and Conflict Transformation in the Great Lakes Region of Africa. Bloomington, IN: Trafford Publishing, 45-72. 287

Ngaruko, Floribert & Nkurunziza, Janvier. 2005. Civil War and its Duration in Burundi. In Collier & Sambanis (eds.) Understanding Civil War: Evidence and Analysis. WashingtonDC, MD: World Bank, 35-62. Nimuraba, Sixte Vingy. 2014. Establishing a Truth and Reconciliation Commision in Burundi: Perspectives on Possibilities and Challenges. Unpublished MsC Thesis: George Mason University. Nindorera, Agnes. 2003. Ubishingantahe as a Base for Political Transformation in Burundi. Consortium of Gender, Security and Human Rights Working Paper 102 (unpublished). Available online at http://genderandsecurity.org/sites/default/files/ubushingantahe_as_a_base_for_political_trans formation_in_burundi_2.pdf Nkundabagenzi, F. 1958. Voici le détail historique du règne des Banyiginya au Rwanda.. Rwanda politique, 35-36. Nkurikiyimfura, Jean-Népomucène. 1994. Le gros bétail et la société rwandaise: Évolution historique des XIIe-XIVe siècles à 1958. Paris: Harmattan. Nsanze, Augustin. 2001. Le Burundi Ancien: L’économie du pouvoir de 1875 à 1920. Paris: L’Harmattan. Nsanze, Augustin. 2003. Le Burundi contemporain: L’Etat-nation en question (1956-2002). Paris: L’Harmattan. Ntahombaye, Philippe & Nduwayo, Gaspard. 2007. Identity and Cultural Diversity in Conflict Resolution and Democratisation for the African Renaissance: The Case of Burundi. Africal Journal on Conflict Resolution, 7 (2), 239-274. Nuttall, Sarah. 2009. Entanglement: Literary and Cultural Reflections on Post-Apartheid. Johannesburg: Wits University Press. Nyakabwa, Rose Kabahenda. 2002. Statelessness and the Batutsi Refugees’ Invasion of Rwanda 1990-1994. Unpublished PhD thesis: University of London. Nzongola-Ntalaja, Georges. 2002. The Congo from Leopold to Kabila: A People’s History. London: Zed Books. Okosun, T. Y. and Kibiswa, N. 2013. Human Rights Violations and Genocide in the Democratic Republic of the Congo. Contemporary Justice Review, 16 (4), 482-493. Omobowale, Ayokunle. 2015. Stories of the ‘Dark’ Continent: Crude Constructions, Diasporic Identity, and International Aid to Africa. International Sociology Review, 30 (2), 108-118. Oppenheimer, Joshua (dir.). 2014. The Look of Silence [documentary film]. Copenhagen: Final Cut for Real DK. Osterhammel, Jürgen. 2005. Colonialism: A Theoretical Overview (2nd ed.), trans. Shelley L. Frisch. Princeton, NJ: Markus Wiener Publishers. Osterhammel, Jürgen. 2014. The Transformation of the World: A Global History of the Nineteenth Century, trans. Patrick Camiller. Princeton, NJ: Princeton University Press. Outhwaite, William. 1987. New Philosophies of Social Science: Realism, Hermeneutics and Critical Theory. Basingstoke: Palgrave Macmillan. Outhwaite, William. 2007. Hermeneutic and Phenomenological Approaches. In Turner & Risjord (eds.) Philosophy of Anthropology and Sociology. Amsterdam: Elsevier. Outhwaite, William. 2008. European Society. Cambridge: Polity. Outhwaite, William. 2008a. Hans-Georg Gadamer. In Skinner (ed.) The Return of Grand Theory in the Human Sciences. Cambridge: Canto, 23-39. Outhwaite, William, 2014. Hermeneutics and the Social Sciences. In Malpas & Gander (eds.) The Routledge Companion to Hermeneutics. London: Routledge, 486-498. Palmer, Jack. 2015. ‘Connected Sociologies’, Colonialism and Genocide. Paper presented at ‘Reconstructing Social Theory, History and Practice’, Meeting of the International Social Theory Consortium, 17th – 19th June 2015, University of Cambridge. 288

PARMEHUTU. 1960, September 5th. Communique Le M.D.R PARMEHUTU - Ruanda. Gitarama: Mouvement Democratique Republicain-PARMEHUTU. Pearce, Carol. 1989. African Philosophy and the Sociological Thesis. Philosophy of the Social Sciences, 22 (4), 440-460. Peirce, C. S. 1998 [1903]. Pragmatism as the Logic of Abduction. In Houser (ed.) The Essential Peirce. Bloomington, IN: Indiana University Press, 226-241. Pinker, Steven. 2011. The Better Angels of our Nature: The Decline of Violence in History and its Causes. London: Penguin. Platt, Jennifer. 1981. Evidence and Proof in Documentary Research. The Sociological Review, 29 (1), 31-66. Polt, Richard. 1999. Martin Heidegger: An Introduction. Ithaca, NY: Cornell University Press. Pottier, Johan. 2002. Re-Imagining Rwanda: Conflict, Survival and Disinformation in the Late Twentieth Century. Cambridge: Cambridge University Press. Powell, Christopher. 2011. Barbaric Civilization: A Critical Sociology of Genocide. Montreal: McGill-Queen’s University Press. Prunier, Gérard. 1995. The Rwanda Crisis: History of a Genocide. London: Hurst & Company. Prunier, Gérard. 1995a. Burundi: Descent into Chaos or a Manageable Crisis? Refugee Survey Quarterly, 14 (1-2), 128-171. Prunier, Gérard. 2009. From Genocide to Continental War: The ‘Congolese’ Conflict and the Crisis of Contemporary Africa. London: Hurst & Company. Quijano, Anibal. 2008. Coloniality of Power, Eurocentrism, and Social Classification. In Moraña, Dussel & Jáuregui (eds.) Coloniality at Large: Latin America and the Postcolonial Debate. Durham, NC: Duke University Press, 181-225. Radio Rwanda. 1990, October 21st. Interview Granted by Professor Anastase Gasana to Bamwanga, a Journalist at Radio Rwanda. Transcript available online at http://migs.concordia.ca/links/Rwandan_Radio_Transcripts_RadioRwanda.htm Randeria, Shalini & Conrad, Sebastian (eds.). 2002. Jenseits des Eurozentrismus: postkoloniale Perspektiven in den Geschichts und Kulturwissenschaften. Frankurt: Campus Verlag. Randeria, Shalini. 2006. Entangled Histories: Civil Society, Caste Solidarities and Legal Pluralism in Post-colonial India. In Keane (ed.) Civil Society: Berlin Perspectives. New York, NY: Berghahn Books. Ranger, Terence. 1982. The Invention of Tradition in Colonial Africa. In Hobsbawm & Ranger (eds.) The Invention of Tradition. Cambridge: Cambridge University Press, 211-262. Ray, Larry. 1999. ‘Fundamentalism’, Modernity, and the New Jacobins. Economy and Society, 28 (2), 198-221. Ray, Larry. 2011. Violence and Society. London: SAGE. Ray, Larry. 2013. Mark of Cain: Shame, Desire and Violence. European Journal of Social Theory, 16 (3), 292-309. Reed, Cyrus. 1996. Exile, Reform, and the Rise of the Rwandan Patriotic Front. The Journal of Modern African Studies, 34 (3), 479-501. Reid, Richard. 2011. Past and Presentism: The ‘Precolonial’ and the Foreshortening of African History. Journal of African History, 52 (2), 135-155. Renton, David; Seddon, David & Zeilig, Leo. 2007. The Congo: Plunder and Resistance. London: Zed Books Ltd. Reyntjens, Filip. 1986. La deuxième république Rwandaise: évolution, bilan et perspectives. Afrika Focus, 2 (3-4). 273-298. Reyntjens, Filip. 1987. Chiefs and Burgomasters in Rwanda: The Unfinished Quest for a Bureaucracy. The Journal of Legal Pluralism and Unofficial Law, 19 (25-26), 71-97. Reyntjens, Filip. 1996. Rwanda: Genocide and Beyond. Journal of Refugee Studies. 9 (3), 240-251. 289

Reyntjens, Filip. 1999. The Second Congo War: More than a Remake. African Affairs, 98 (391), pp. 241-250. Reyntjens, Filip. 2016. (Re-)imagining a Reluctant Post-Genocide Society: The Rwandan Patriotic Front’s Ideology and Practice. Journal of Genocide Research, 18 (1), 61-81. Ricoeur, Paul. 1967. The Symbolism of Evil, trans. Buchanan. Boston, MA: Beacon. Ricoeur, Paul. 1976. History and Hermeneutics. The Journal of Philosophy, 73 (19), 683-695. Ricoeur, Paul. 1980. Narrative Time. Critical Inquiry, 7 (1), 169-190. Ricoeur, Paul. 1988. Time and Narrative, Vol. 3, trans. Blamey & Pellauer. Chicago, IL: University of Chicago Press. Ricoeur, Paul. 1991. From Text to Action: Essays in Hermeneutics II, trans. Blamey & Thompson. Evanston, IL: Northwestern University Press. Rodney, Walter. 1972. How Europe Underdeveloped Africa. London: Bogle-L’Ouverture Publications. Rodríguez, E., Boatcă, M. and Costa, S. 2010. Decolonizing European Sociology: Transdisciplinary Approaches. Surrey: Ashgate. Rose, Gillian. 1996. Mourning Becomes the Law: Philosophy and Representation. Cambridge: Cambridge University Press. Rosenberg, Sheri. 2012. Genocide is a Process, Not an Event. Genocide Studies and Prevention: An International Journal, 7 (1), 16-23. Rothberg, Michael. 2009. Multidirectional Memory: Remembering the Holocaust in the Age of Decolonization. Stanford, CA: Stanford University Press. Rothermund, Dietmar. 2006. The Routledge Companion to Decolonization. London: Routledge. RTLM. 1994, April 11th. Broadcast of Radio Télévision Libre des Mille Collines, April 11th (transcribed and translated by Charles Zikuliza, 3rd September 1995). Available online at: http://migs.concordia.ca/links/documents/RTLM_11Apr94_eng_K014-6083-K014-6089.pdf Rueschemeyer, Dietrich. 1984. Theoretical Generalization and Historical Particularity in the Comparative Sociology of Reinhard Bendix. In Skocpol (ed.) Vision and Method in Historical Sociology. Cambridge: Cambridge University Press, 129-169. Russell, Aidan. 2015. Obedience and Selective Genocide in Burundi. Africa, 8 (3), 437-456. Russell, Aidan. 2015a. Rebel and Rule in Burundi, 1972. International Journal of African Historical Studies, 48 (1). 73-97. Russell, Bertrand. 1964, February 6th. L’Extermination des Tutsi. Le Monde. Rutake, Pascal & Gahama, Joseph. 1998. Ethnic Conflict in Burundi. In Nnoli (ed.) Ethnic Conflicts in Africa. Dakar: CODESRIA, 79-104. Rutanga, Murindwa. 1991. Nyabingi Movement: People’s Anti-colonial Struggles in Kigezi, 1910- 1930. Kigezi: CBR Working Paper. Rwabukumba, Joseph & Mundandagizi, Vincent. 1974. Les formes historiques de la dépedence personelle dans l’État rwandais. Cahiers d’Etudes Africaines, 14 (53), 6-25. Ryckmans, Pierre. 1948. Dominer Pour Servir. Paris: Universelle. Ryckmans, Pierre. 1953. Une page d'histoire coloniale: l'occupation allemande dans l'Urundi. Institut royal colonial belge. Sagall, Sabby. 2013. Final Solutions: Human Nature, Capitalism and Genocide. London: Pluto. Said, E. W. 1978. Orientalism: Western Conceptions of the Orient. London: Penguin Books. Sanders, Edith. 1969. The Hamitic Hypothesis: Its Origins and Functions in Time Perspective. The Journal of African History, 10 (4), 521-532. Sartre, Jean-Paul. 1965. Preface. In Fanon (auth.) The Wretched of the Earth, trans. Constance Farrington. London: Macgibbon & Kee, 7-34. Savage, Rowan. 2007. ‘Disease Incarnate’: Biopolitical Discourse and Genocidal Dehumanisation in the Age of Modernity. Journal of Historical Sociology, 20 (3), 404-440. 290

Sayer, Andrew. 2010. Reductionism in Social Science. In Lee (ed.) Questioning Nineteenth-Century Assumptions about Knowledge, II: Reductionism. Albany, NY: SUNY Press, 5-39. Sayyid, S. 2015. Towards a Critique of Eurocentrism: Remarks on Wittgenstein, Philosophy and Racism. In Araújo & Maeso (eds.) Eurocentrism, Racism and Knowledge: Debates on History and Power in Europe and the Americas. Basingstoke: Palgrave Macmillan, 80-92. Schabas, William. 2000. Genocide in Internatonal Law. Cambridge: Cambridge University Press. Schabas, William. 2000a. Hate Speech in Rwanda: The Road to Genocide. McGill Law Journal, 46 (1), 141-171. Schäfer, Wolf. 2014. Reconfiguring Area Studies for the Global Age. In Arjomand (ed.) Social Theory and Regional Studies in the Global Age. Albany, NY: SUNY Press, 145-175. Schaller, Dominik. 2005. Raphael Lemkin’s View of European Colonial Rule in Africa: Between Condemnation and Admiration. Journal of Genocide Research, 7 (4), 531-538. Scheper-Hughes, Nancy & Bourgois, Phillipe (eds.). 2004. Violence in War and Peace: An Anthology. Oxford: Blackwell. Scherrer, Christian. 1999. Towards a Theory of Modern Genocide: Comparative Genocide Research; Definition, Criteria, Typologies, Cases, Key Elements, Patterns and Voids. Journal of Genocide Research, 1 (1), 13-23. Schiel, Tilman. 2005. Modernity, Ambivalence and the Gardening State. Thesis Eleven, 83 (1), 78- 89. Schmidt, Lawrence. 2006. Understanding Hermeneutics. Durham: Acumen. Schneiderhan, Erik. 2013. Genocide Reconsidered: A Pragmatist Approach. Journal for the Theory of Social Behaviour, 43 (3), 280-300. Schoenbrun, D. L. 1998. A Green Place, A Good Place: Agrarian Change, Gender, and Social Identity in the Great Lakes Region to the 15th Century. Portsmouth, NH: Heinemann. Scholz, Anton. 2015. Hutu, Tutsi, and the Germans: Racial Cognition in Rwanda under German Colonial Rule. Unpublished Masters Thesis: Leiden University. Schraube, Ernst. 2005. ‘Torturing Things until they Confess’: Günther Anders’ Critique of Technology. Science as Culture, 14 (1), 77-85. Schutz, Alfred. 1967. The Phenomenology of the Social World. Evanston: Northwestern University Press. Scott, James C. 1990. Domination and the Arts of Resistance: Hidden Transcripts. New Haven, CT: Yale University Press. Scott, James C. 1999. Seeing like a State: How Certain Schemes to Improve the Human Condition have Failed. New Haven, CT: Yale University Press. Scott, John. 1990. A Matter of Record: Documentary Sources in Social Research. Cambridge: Polity. Segal, Aaron. 1964. Massacre in Rwanda. Fabian Research Series 240. London: Fabian Society. Seligman, Charles. 1966. Races of Africa (4th ed.). Oxford: Oxford University Press. Semelin, Jacques. 2007. Purify and Destroy: The Political Uses of Massacre and Genocide. London: Hurst. Seth, Sanjay. 2009. Historical Sociology and Postcolonial Theory: Two Strategies for Challenging Eurocentrism. International Political Sociology, 3(3), 334-338. Seth, Sanjay. 2016. Is Thinking with ‘Modernity’ Eurocentric? Cultural Sociology [online first version], 1-14. Available online at: http://cus.sagepub.com/content/early/2016/03/30/1749975516637203.abstract Shantz, H. L. 1922. Urundi, Territory and People. The Geographical Review, 12 (1), 329-357. Shaw, Martin. 2002. Globality and Historical Sociology: State, Revolution and War Revisited. In Hobden and Hobson (eds.) Historical Sociology of International Relations. Cambridge: University of Cambridge Press, 82-98. Shaw, Martin. 2003. War and Genocide: Organized Killing in Modern Society. Cambridge: Polity. Shaw, Martin. 2007. What is Genocide? Cambridge: Polity. 291

Shaw, Martin. 2007a. The General Hybridity of War and Genocide. Journal of Genocide Research, 9 (3), 461-473. Shaw, Martin. 2011. Britain and Genocide: Historical and Contemporary Parameters of National Responsibility. Review of International Studies 37, 2417-2438. Shaw, Martin. 2013. Genocide and International Relations: Changing Patterns in the Transitions of the Late Modern World. Cambridge: Cambridge University Press. Shilliam, Robbie (ed.). 2011. International Relations and Non-Western Thought: Imperialism, Colonialism, and Investigations of Global Modernity. London: Routledge. Shils, Edward. 1958. Tradition and Liberty: Antinomy and Interdependence. Ethics, 68 (3), 153-165. Shils, Edward. 1981. Tradition. Chicago, IL: University of Chicago Press. Shyaka, Anastase. 2006. Understanding the Conflicts in the Great Lakes Region: An Overview. Journal of African Conflicts and Peace Studies, 1 (1), 5-12. Silber, Ilana. 2011. Deciphering Transcendence and the Open Code of Modernity: S. N. Eisenstadt’s Comparative Hermeneutics of Civilizations. Journal of Classical Sociology, 11 (3), 269-280. Silverman, Max. 2013. Palimpsestic Memory: The Holocaust and Colonialism in French and Francophone Fiction and Film. New York, NY: Berghahn Books. Simmel, Georg. 2009 [1908]. Sociology: Inquiries into the Construction of Social Forms, vol I. trans. Blasi, Jacobs & Kanjirathinkal. Leiden: Brill. Simms, Karl. 2003. Paul Ricoeur. London: Routledge. Skocpol, Theda. 1979. States and Social Revolutions: A Comparative Analysis of France, Russia, and China. Cambridge: Cambridge University Press. Skocpol, Theda, 1984. Sociology’s Historical Imagination. In Skocpol (ed.) Vision and Method in Historical Sociology. Cambridge, MA: Cambridge University Press, 1-21. Skonhoft, C. G. 2000. ‘Why Should I Send my Child to School?’ A Study of Burundian Hutu Refugees’ Experiences of Exclusion from Education and how this Motivates Education in Tanzanian Exile. Norsk Geografisk Tidsskrift – Norwegian Journal of Geography, 54 (1), 116-121. Smith, Dennis. 1999. Zygmunt Bauman: Prophet of Post-Modernity. Cambridge: Polity. Smith, Iain & Stucki, Andreas. 2011. The Colonial Development of Concentration Camps (1868- 1902). The Journal of Imperial and Commonwealth History, 39 (3), 417-437. Smith, Jeremy. 2015. Grounds for Engagement: Dissonances and Overlaps at the Intersection of Contemporary Civilizations Analysis and Postcolonial Sociology. Current Sociology, 63 (4), 566-585. Speke, J. H. 1864. Journal of the Discovery of the Source of the Nile (2nd ed.). London: William Blackwood and Sons. Speke, J. H. & Burton, Richard. 1858. A Coasting Voyage from Mombasa to the Pangani River; Visit to Sultan Kimwerel; And Progress of the Expedition into the Interior. Journal of the Royal Geographical Society of London, 28 (1), 188-226. Spencer, Philip. 2012. Genocide since 1945. London: Routledge. Spivak, G.C. 1988. Can the Subaltern Speak? In Nelson and Grossberg (eds.) Marxism and the Interpretation of Culture Chicago, IL: University of Illinois Press, 271-313. Spohn, Willfried. 2011. An Appraisal of Schmuel Noah Eisestadt’s Global Historical Sociology. Journal of Classical Sociology, 11 (3), 281-301. Squitieri, Tom. 1993, 27th December. Africa’s Bloodiest War – Burundi war toll 200,000 since Oct. 21. USA Today, 1a. Stanton, Gregory. 2013. The Ten Stages of Genocide. Genocide Watch. Obtained online at http://www.genocidewatch.org/images/Ten_Stages_of_Genocide_by_Gregory_Stanton.pdf Steinmetz, George. 2007. The Devil’s Handwriting: Precoloniality and the German Colonial State in Qingdao, Samoa and Southwest Africa. Chicago, IL: University of Chicago Press. 292

Steinmetz, George. 2008. The Colonial State as a Social Field: Ethnographic Capital and Native Policy in the German Overseas Empire before 1914. American Sociological Review, 73 (1), 589-612. Steinmetz, George (ed.). 2013. Sociology and Empire: The Imperial Entanglements of a Discipline. Durham, NC: Duke University Press. Steinmetz, George. 2014. The Sociology of Empires, Colonies, and Postcolonialism. Annual Review of Sociology, 40 (1), 77-103. Stewart, Pamela & Strathern, Andrew. 2002. Violence: Theory and Ethnography. London: Continuum. Stone, Dan. 1999. Modernity and Violence: Theoretical Reflections on the Einzatsgruppen. Journal of Genocide Research, 1 (3), 367-378. Stone, Dan. 2004. Genocide as Transgression. European Journal of Social Theory, 7 (1), 45-65. Stone, Dan. 2004a. The Historiography of Genocide: Beyond ‘Uniqueness’ and Ethnic Competition. Rethinking History: The Journal of Theory and Practice, 8 (1), 127-142. Storey, Andy. 2001. Structural Adjustment, State Power & Genocide: The World Bank and Rwanda. Review of African Political Economy, 28 (89), 365-385. Stråth, Bo. 2016. Europe’s Utopias of Peace:1816, 1919, 1951. London: Bloomsbury. Straus, Scott. 2001. Contested Meanings and Conflicted Imperatives: A Conceptual Analysis of Genocide. Journal of Genocide Research, 3 (3), 349-375. Straus, Scott. 2004. How Many Perpetrators were there in the Rwandan Genocide? An Estimate. Journal of Genocide Research, 6 (1), 85-98. Straus, Scott. 2008. The Order of Genocide: Race, Power, and War in Rwanda. New York, NY: Cornell University Press. Strizek, Helmut. 2006. Geschenkte Kolonien: Ruanda und Burundi unter deutsch Herrschaft. Berlin: Christoph Links Verlag. Suzuki, Shogo. 2009. Civilizational and Empire: China and Japan’s Encounter with European International Society. London: Routledge. Szakolczai, Arpad. 2000. Reflexive Historical Sociology. London: Routledge. Szakolczai, Arpad & Wydra, Harald. 2006. Contemporary East Central European Social Theory. In Delanty (ed.) Handbook of Contemporary European Social Theory. London: Routledge, 138- 152. Taïwo, Olúfémi. 2010. How Colonialism Preempted Modernity in Africa. Indianapolis, IN: Indiana University Press. Tatum, Dale. 2010. Genocide at the Dawn of the 21st Century: Rwanda, Bosnia, Kosovo, and Darfur. Basingstoke: Palgrave Macmillan. Taylor, Charles. 2002. Gadamer on the Human Sciences. In Dostal (ed.) The Cambridge Companion to Gadamer. Cambridge: Cambridge University Press, 126-142. Taylor, Charles. 2004. Modern Social Imaginaries. Durham, NC: Duke University Press. Taylor, Christopher. 1988. The Concept of Flow in Rwandan Popular Medicine. Social Science & Medicine, 27 (12), 1343-1348. Taylor, Christopher. 1999. Sacrifice as Terror: The Rwandan Genocide of 1994. Oxford: Berg. Taylor, Christopher. 2004. Dual Systems in Rwanda: Have They Ever Really Existed? Anthropological Theory, 4 (3), 353-371. Taylor, Christopher. 2011. Molders of Mud: Ethnogenesis and Rwanda’s Twa. Ethnos: Journal of Anthropology, 76 (2), 183-208. Tempels, Placide. 1959 [1945]. Bantu Philosophy, trans. E. Possoz, and Margaret Read. Paris: Présence africaine. Tester, Keith. 1997. Moral Culture. London: SAGE. Tester, Keith. 2004. The Social Thought of Zygmunt Bauman. Basingstoke: Palgrave Macmillan. 293

Therborn, Göran. 1995. European Modernity and Beyond: The Trajectory of European Societies, 1945-2000. London: SAGE. Therborn, Göran. 2000. ‘Modernisation’ Discourses, their Limitations, and their Alternatives. In Schelkle & Krauth (eds.) Paradigms of Social Change: Modernisation, Development, Transforamtion, Evolution. Frankfurt: Campus Verlag, 49-72. Therborn, Göran. 2003. Entangled Modernities. European Journal of Social Theory, 6 (1), 293-305. Therborn, Göran. 2010. Different Roads to Modernity and Their Consequences: A Sketch. In Rodríguez, Boatcă & Costa (eds.) Decolonizing European Sociology: Transdisciplinary Approaches. Farnham: Ashgate, 71-84. Therborn, Göran. 2011. The World: A Beginner’s Guide. Cambridge: Polity. Therborn, Göran. 2014. From Civilizations to Modernity: Divisions and Connections of the World, and Their Legacy – A Historical Social Geology. In Arjomand (ed.) Social Theory and Regional Studies for the Global Age. Albany, NY: SUNY Press, 267-290. Thibon, Christian. 2004. Histoire démographique du Burundi. Paris: Karthala. Thiselton, Anthony. 2009. Hermeneutics: An Introduction. Cambridge: Eerdmans. Thomas, Lynn. 2011. AHR Roundtable: Modernity’s Failings, Political Claims, and Intermediate Concepts. American Historical Review, 116 (3), 727-740. Thomas, Martin et al (eds.). 2008. Crises of Empire: Decolonization and Europe’s Imperial States, 1918-1975. London: Bloomsbury. Thompson, Allan (ed.). 2007. The Media and the Rwandan Genocide. London: Pluto Press. Thomson, Susan; Ansoms, An & Murison, Jude. 2013. Emotional and Ethical Challenges for Field Research in Africa: the Story Behind the Findings. Basingstoke: Palgrave Macmillan. Tiryakian, Edward. 1991. Modernisation: Exhumetur in Pace. International Sociology, 6 (2), 165- 180. Tiryakian, Edward. 2011. A Sociological Odyssey: The Comparative Voyage of S. N. Eisenstadt. Journal of Classical Sociology, 11 (3), 241-250. Todorov, Tzvetan. 1992 [1982]. The Conquest of America, trans. R. Howard. New York, NY: Harper-Perennial. Totten, Samuel & Bartrop, Paul. (eds.). 2009. The Genocide Studies Reader. London: Routledge. Totten, Samuel; Parsons, William & Charny, Israel. (eds.). 2009. Century of Genocide: Critical Essays and Eyewitness Accounts. London: Routledge, Touraine, Alain. 1989. Is Sociology Still the Study of Society? Thesis Eleven, 23 (1), 5-34. Traverso, Enzo. 2003. The Origins of Nazi Violence. New York, NY: The New Press. Travis, Hanibal. 2016. Counterinsurgency as Genocidal Intent: From the Ottoman Christians to the Bosnian Muslims. In Demirdjian (ed.) The Armenian Genocide Legacy. Basingstoke: Palgrave Macmillan, 149-167. Trouwborst, Albert. 2008. The End of the Early State: The Case of Burundi. Social Evolution and History, 7 (1), 166-170. Turner, B. S. 2006. Classical Sociology and Cosmopolitanism: A Critical Defence of the Social. British Journal of Sociology, 57 (1), 133-151. Turner, Simon. 1998. Representing the Past in Exile: The Politics of National History among Burundian Refugees. Refuge, 17 (6), 22-29. Twagilimana, Aimable. 2016. Historical Dictionary of Rwanda, 2nd ed. Lanham, MD: Rowman & Littlefield. Twagiramutara, Pancrace. 1998. Ethnicity and Genocide in Rwanda. In Nnoli (ed.) Ethnic Conflicts in Africa. Dakar: CODESRIA, 105-130. United Nations. 1945. Charter of the United Nations and Statute of the International Court of Justice. San Francisco. 294

United Nations. 1957. United Nations Visisting Mission to Trust Territories in East Africa, 1957: Report on Ruanda-Urundi Together with the Relevant Resolution of the Security Council. New York, NY: United Nations. 1978. Rwanda, Burundi and Zaire: Convention establishing the Economic Community of the Great Lakes Countries (CEPGL). Concluded at Gisenyi on 20 September 1976. United Nations. 2013, February 24. Peace, Security and Cooperation Framework for the Democratic Republic of Congo and the Region. Addis Ababa. United Nations High Commissioner for Human Rights, Office of the High Commissioner for Human Rights. 2010. Report of the mapping exercise documenting the most serious violations of human rights and international humanitarian law committed within the territory of the Democratic Republic of the Congo between March 1993 and June 2003. Retrieved from http://www.ohchr.org/EN/Countries/AfricaRegion/Pages/RDCProjetMapping.aspx Uvin, Peter. 1997. Prejudice, Crisis, and Genocide in Rwanda. African Studies Review, 40 (2), 91- 115. Uvin, Peter. 1998. Aiding Violence: The Development Enterprise in Rwanda. West Hartford, CT: Kumarian Press. Uvin, Peter. 1999. Ethnicity and Power in Burundi and Rwanda: Different Paths to Mass Violence. Comparative Politics, 31 (3), 253-271. Uvin, Peter. 2004. On Counting, Categorizing, and Violence in Burundi and Rwanda. In Kertzer and Arel (eds.) Census and Identity: The Politics of Race, Ethnicity, and Language in National Censuses. Cambridge: Cambridge University Press.148-175. Uwimbabazi, Penine & Lawrence, Ralph. 2013. Indigenous Practice, Power and Social Control: The Paradox of the Practice of Umuganda in Rwanda. Alternation: Interdisciplinary Journal for the Study of the Arts and Humanities in Southern Africa, 20 (1), 248-272. Vandenberghe, Frédéric. 2009 [1997]. A Philosophical History of German Sociology. London: Routledge. Vandenginste, Stef. 2014. Governing Ethnicity after Genocide: Ethnic Amnesia in Rwanda versus Ethnic Power-Sharing in Burundi. Journal of Eastern African Studies, 8 (2), 263-277. Vansina, Jan. 1985. Oral Tradition as History. Madison, WI: University of Wisconsin Press. Vansina, Jan. 2004. Antecedents to Modern Rwanda: The Nyiginya Kingdom. Madison, WI: University of Wisconsin Press. Vanthemsche, Guy. 2012. Belgium and the Congo, 1885-1980, trans. Cameron & Windross. Cambridge: Cambridge University Press. Varcoe, Ian. 1998. Identity and the Limits of Comparison: Bauman’s Reception in Germany. Theory, Culture & Society, 15 (1), 57-72. Veracini, Lorenzo. 2010. Settler Colonialism: A Theoretical Overview. Basingstoke: Palgrave Macmillan. Verwimp, Philip. 2000. Development Ideology, the Peasantry and Genocide: Rwanda Represented in Habyarimana’s Speeches. Journal of Genocide Research, 2 (3), 325-361. Verwimp, Philip. 2006. Machetes and Firearms: The Organization of Massacres in Rwanda. Journal of Peace Research, 43 (1), 5-22. Verwimp, Philip. 2013. Peasants in Power: The Political Economy of Development and Genocide in Rwanda. New York, NY: Springer. Vidal, Claudine. 1974. Économie de la société féodale rwandaise (The Economics of Rwandan Feudal Society). Cahiers d'études africaines, 14 (53), 52-74. Vlassenroot, Koen & Romkema, Hans. 2002. The Emergence of a New Order? Resources and War in Eastern Congo. The Journal of Humanitarian Assistance, available online at http://www.jha.ac/articles/a111.htm 295

Wagner, Michele. 1993. Trade and Commercial Attitudes in Burundi before the Nineteenth Century. The International Journal of African Historical Studies, 26 (1), 149-166. Wagner, Peter. 1994. A Sociology of Modernity: Liberty and Discipline. London: Routledge. Wagner, Peter. 2001. Modernity: One or Many? In Blau (ed.) The Blackwell Companion To Sociology. Oxford: Blackwell, 30-32. Wagner, Peter. 2008. Modernity as Experience and Interpretation: A New Sociology of Modernity. Cambridge: Polity. Wagner, Peter. 2009. Modernity as Experience and as Interpretation: Towards Something Like a Cultural Turn in the Sociology of ‘Modern Society’. In Hedström & Knöbl (eds.) Frontiers of Sociology. Leiden: Brill, 247-266. Wagner, Peter. 2010. Successive Modernities and the Idea of Progress: A First Attempt. Distinktion: Scandinavian Journal of Social Theory, 11 (2), 9-24. Wagner, Peter. 2010a. Multiple Trajectories of Modernity: Why Social Theory needs Historical Sociology. Thesis Eleven, 100 (1), 53-60. Wagner, Peter. 2012. Modernity: Understanding the Present. Cambridge: Polity. Wagner, Peter. 2014. World-Sociology beyond the Fragments: Oblivion and Advance in the Comparative Analysis of Modernities. In Arjomand (ed.) Social Theory and Regional Studies in the Global Age. Albany, NY: SUNY Press, 293-312. Wagner, Peter. 2015. Introduction. In Wagner (ed.) African, American and European Trajectories of Modernity: Past Oppression, Future Justice? Edinburgh: Edinburgh University Press, 1-12. Wallerstein, Immanuel. 1973. Africa in a Capitalist World. Issue: A Journal of Opinion. 3 (3), 1-11. Wallerstein, Immanuel. 1974. Dependence in an Interdependent World: The Limited Possibilities of Transformation within the Capitalist World Economy. African Studies Review, 17 (1), 1-26. Wallerstein, Immmanuel. 2004. World Systems Analysis: An Introduction. Durham, NC: Duke University Press. Watson, Catherine. 1991. Exile from Rwanda: Background to an Invasion. Washington DC, MD: US Committee on Refugees. Watt, Nigel. 2008. Burundi: Biography of a Small African Country. London: Hurst. Weber, Max. 1978. Economy and Society: An Outline of Interpretive Sociology, edited by Guenther Roth & Claus Wittich. Los Angeles, CA: University of California Press. Weber, Max. 1993 [1922]. The Sociology of Religion, trans. Fischoff. Boston, MA: Beacon Press. Weber, Max. 2001 [1905]. The Protestant Ethic and the Spirit of Capitalism, trans. Parsons. London: Routledge. Weber, Max. 2008 [1919]. Science as a Vocation. In Dreijmanis (ed.) Max Weber’s Complete Writings on Academic and Political Vocations, trans. Wells. New York, NY: Algora Publishing, 25-52. Webster, J. B. 1979. Chronology, Migration and Drought in Interlacustrine Africa. London: Longman & Dalhousie University Press. Weinstein, Warren. 1965. Racial Peace and Royalty. Africa Today, 12 (6), 12-15. Weitz, Eric. 2003. A Century of Genocide: Utopias of Race and Nation. Princeton, NJ: Princeton University Press. Werner, Michael & Zimmerman, Benedicte. 2006. Beyond Comparison: Histoire Croisee and the Challenge of Reflexivity. History and Theory, 45 (1), 30-50. White, Hayden. 1984. The Question of Narrative in Contemporary Historical Theory. History and Theory, 23 (1), 1-33. Wieviorka, Michel. 2012. Evil, trans Kristin Couper-Lobel. Cambridge: Polity. Wimmer, Andreas. 2002. Nationalist Exclusion and Ethnic Conflict: Shadows of Modernity. Cambridge: Cambridge University Press. Wimmer, Andreas and Glick-Schiller, Nina. 2002. Methodological Nationalism and Beyond: Nation–State Building, Migration and the Social Sciences. Global networks, 2 (4), 301-334. 296

Wolfe, Patrick. 2006. Settler Colonialism and the Elimination of the Native. Journal of Genocide Research, 8 (4), 387-409. Wolff, Kurt. 2002. What it Contains. Lanham, MD: Lexington Books. Wood, Sophia. 2010. Film and Atrocity: The Holocaust as Spectacle. In Wilson and Crowder- Taraborrelli (eds.) Film and Genocide. Madison, WI: University of Wisconsin Press, 21-44. Wrangham, Richard & Peterson, Dale. 1996. Demonic Males: Apes and the Origins of Human Violence. New York, NY: Mariner Books. Young, Crawford. 1994. The African Colonial State in Comparative Perspective. New Haven, CT: Yale University Press. Young, Crawford. 2008. Zaire, Rwanda and Burundi. In Fage & Oliver (eds.) The Cambridge History of Africa, vol. 8 (1940-1975). Cambridge: Cambridge University Press, 698-754. Young, Crawford. 2015. Imperial Endings and Small States: Disorderly Decolonization for the Netherlands, Belgium, and Portugal. In Jerónimo & Pinto (eds.) The Ends of European Colonial Empires: Cases and Comparisons. Basingstoke: Palgrave Macmillan, 101-125. Zimmerer, Jürgen. 2005. The Birth of the Ostland out of the Spirit of Colonialism: a Postcolonial Perspective on the Nazi Policy of Conquest and Extermination. Patterns of Prejudice, 39 (2), 197-219. Zimmerer, Jurgen; Perraudin, Michael & Heady, Kate (eds.). 2011. German Colonialism and National Identity. London: Routledge. Žižek, Slavoj. 2009. Violence: Six Sideways Reflections. London: Profile Books.

297